Skip to content
This repository has been archived by the owner on Oct 11, 2024. It is now read-only.

Commit

Permalink
increase timeouts (#253)
Browse files Browse the repository at this point in the history 
SUMMARY:
* increase "RELEASE" workflow timeouts to 12 hours

TEST PLAN:
will cherry pick and trigger job manually on release, `v0.3.0`

Co-authored-by: andy-neuma <[email protected]>
(cherry picked from commit 93183d6)
  • Loading branch information
@andy-neuma
andy-neuma authored and andy-neuma committed May 20, 2024
1 parent 9a97d84 commit 707295f
Showing 1 changed file with 8 additions and 8 deletions.
16 changes: 8 additions & 8 deletions .github/workflows/release.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,12 +26,12 @@ jobs:

test_label_solo: aws-avx2-32G-a10g-24G
test_label_multi: aws-avx2-192G-4-a10g-96G
test_timeout: 480
test_timeout: 720
test_skip_list: neuralmagic/tests/skip-for-release.txt

benchmark_label: aws-avx2-32G-a10g-24G
benchmark_config_list_file: ./.github/data/nm_benchmark_nightly_configs_list.txt
benchmark_timeout: 180
benchmark_timeout: 720
push_benchmark_results_to_gh_pages: ${{ inputs.push_benchmark_results_to_gh_pages }}
secrets: inherit

Expand All @@ -44,12 +44,12 @@ jobs:

test_label_solo: aws-avx2-32G-a10g-24G
test_label_multi: aws-avx2-192G-4-a10g-96G
test_timeout: 480
test_timeout: 720
test_skip_list: neuralmagic/tests/skip-for-release.txt

benchmark_label: aws-avx2-32G-a10g-24G
benchmark_config_list_file: ./.github/data/nm_benchmark_nightly_configs_list.txt
benchmark_timeout: 180
benchmark_timeout: 720
push_benchmark_results_to_gh_pages: ${{ inputs.push_benchmark_results_to_gh_pages }}
secrets: inherit

Expand All @@ -62,12 +62,12 @@ jobs:

test_label_solo: aws-avx2-32G-a10g-24G
test_label_multi: aws-avx2-192G-4-a10g-96G
test_timeout: 480
test_timeout: 720
test_skip_list: neuralmagic/tests/skip-for-release.txt

benchmark_label: aws-avx2-32G-a10g-24G
benchmark_config_list_file: ./.github/data/nm_benchmark_nightly_configs_list.txt
benchmark_timeout: 180
benchmark_timeout: 720
push_benchmark_results_to_gh_pages: ${{ inputs.push_benchmark_results_to_gh_pages }}
secrets: inherit

Expand All @@ -80,11 +80,11 @@ jobs:

test_label_solo: aws-avx2-32G-a10g-24G
test_label_multi: aws-avx2-192G-4-a10g-96G
test_timeout: 480
test_timeout: 720
test_skip_list: neuralmagic/tests/skip-for-release.txt

benchmark_label: aws-avx2-32G-a10g-24G
benchmark_config_list_file: ./.github/data/nm_benchmark_nightly_configs_list.txt
benchmark_timeout: 180
benchmark_timeout: 720
push_benchmark_results_to_gh_pages: ${{ inputs.push_benchmark_results_to_gh_pages }}
secrets: inherit

2 comments on commit 707295f

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bigger_is_better

Benchmark suite Current: 707295f Previous: d69a34a Ratio
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.8091444987524608 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 235.1887848378199 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 15.999337755546904 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4111.829803175555 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9839566027548535 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 290.12288417761107 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 212.3771931386076 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.454249385616612 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 757.937656928693 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 563.4531186148901 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.0425989902081536 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4185.285330936506 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.074390953724896 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4250.427064182312 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.49189677153223854 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 130.02799258683194 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 121.96088553370323 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4.321490167908952 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1369.437019308668 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 864.1165309947382 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.811267916272564 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 235.46482911543333 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.911496869481959 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4058.597894044245 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.22889213132706 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1521.989968375111 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4.525957328106215 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1434.2306177035784 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1033.1976080796285 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4.179768132436878 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4284.2623357478 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.227905356838155 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3379.479428643248 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 12.196599245713271 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1585.5579019427253 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.47523266044903506 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 125.62300146309794 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 99.48203692066467 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 26.151968039807162 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1699.8779225874657 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 12.17312764254394 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1582.5065935307123 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5.46415595484801 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 710.3402741302413 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.769404508030182 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3863.6396207309367 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 17.559019532849412 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2282.672539270424 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 10.669077828471822 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1386.980117701337 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6.545982864591494 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 850.9777723968942 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11.93794209578227 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1551.9324724516953 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.4501716970141003 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 756.6783579498746 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 562.0661203994387 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - 2:4 Sparse (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5.745263607566038 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - 2:4 Sparse (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2686.255452353577 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9173305255238989 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 119.25296831810687 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.819928597826773 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 366.59071771748046 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 27.581914770154103 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3558.0670053498793 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 15.638160489746495 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4019.007245864849 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.504103112255784 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3849.604896587217 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 22.945464276986225 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2959.9648917312234 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 14.018524218341597 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3602.7607241137903 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.019412890955771 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 132.5236758242502 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.958095165380789 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 384.5523714995026 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6.837314390476668 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 888.8508707619669 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 18.193126584810138 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2365.1064560253176 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 24.492230146228458 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3159.497688863471 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.7687249570220342 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3862.9430809475853 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9167108259452095 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 119.17240737287723 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 8.053085488526209 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4131.232855613945 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.0107814724325914 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4120.091237014381 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9681857570756103 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 288.69040175894855 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 210.0704909985521 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 10.749424441259801 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1397.425177363774 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 26.182563633538525 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1701.866636180004 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.6059272375686353 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1142.682282313125 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 824.591420929737 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.491877158634526 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 130.02280811345062 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 121.9822561603049 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 14.283141483552843 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3670.7673612730805 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4.564189046636082 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1446.345866988508 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1043.416215158777 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.46425725597163137 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 122.72176304354103 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 115.14817967445715 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.7645579942606657 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 99.39253925388654 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 13.987104197589552 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3594.6857787805147 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4.062640274043485 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4164.206280894572 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.3605903155699575 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 729.0132385230847 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 541.7586248770593 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 24.2875307519995 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3133.091467007936 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.4664139195587844 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 126.1245427009561 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 118.18617779006559 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.8464560105260195 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 586.6756991951058 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 413.43750337553365 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.875684099425817 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3972.5762019114622 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 10.129442999862796 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1316.8275899821635 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 25.56470385470706 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1661.7057505559587 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 20.4928006164199 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2664.0640801345867 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - Sparse (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5.745662545485973 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - Sparse (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2686.4419797674213 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - synthetic\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.8759433761833475 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - synthetic\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1488.3622564544055 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.8098249686647738 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 235.27724592642062 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.24101549493706703 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 31.332014341818713 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.502877015738628 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3848.975909073916 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.452271261392815 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3484.383950976824 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.427180777755596 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 445.5335011082275 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.7280020142772654 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 94.64026185604449 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.308546642949134 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 712.9407645865047 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 464.34722788717005 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 49.793731951386206 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3236.5925768401034 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.25710900108094653 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 33.42417014052305 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 23.645870738947142 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3050.317325324181 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.768671525046322 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 1024,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3862.88831317248 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5.713683671598466 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 742.7788773078006 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9839012122894073 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 290.1065521142394 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 212.37507667266857 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9839786295162612 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 290.12937884163466 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 212.391787181085 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 29.28694803850999 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 128,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3778.016296967789 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 26.217236811954624 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1704.1203927770505 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.7630531653413244 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 489.19691149437216 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 48.31851799187511 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 64,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3140.703669471882 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6.843626170941067 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 889.6714022223388 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 20.89127286442713 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 64\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2715.8654723755267 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.234734596410666 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 690.1456362941847 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 513.0086535981611 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.24094037726232453 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 31.32224904410219 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.24104148901896513 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 31.335393572465467 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9481180817991071 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 279.5557768120727 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 204.66708992436725 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.0671400643449758 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4235.569991842855 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.042313183712066 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4184.699713426024 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 8.136747281424158 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Dense (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4174.151355370593 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 13.989302743741435 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 256,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3595.250805141549 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.9955009505415129 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 8\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 259.41512357039664 tokens/s
{"name": "request_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5.746380012037925 prompts/s
{"name": "token_throughput", "description": "VLLM Engine throughput - Dense (with dataset)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"dataset\": \"sharegpt\",\n \"output-len\": 128,\n \"num-prompts\": 1000\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2686.7774384284517 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2.0431944965007474 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2048,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 4186.505523330031 tokens/s
{"name": "request_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7.503787386690349 prompts/s
{"name": "token_throughput", "description": "VLLM Engine prefill throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 512,\n \"output-len\": 1,\n \"num-prompts\": 1,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3849.4429293721487 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.5864307391006403 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1136.5040369136018 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 819.7959021589286 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 0.9175882620939957 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 4\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 119.28647407221945 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.4296742859028595 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 445.8576571673717 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3.425373260563811 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - Dense (synthetic)\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 16\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 445.29852387329544 tokens/s
{"name": "request_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1.7162643550548895 prompts/s
{"name": "input_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 559.7378800526549 tokens/s
{"name": "output_throughput", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 390.08629273171573 tokens/s
{"name": "request_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6.8424535448302395 prompts/s
{"name": "token_throughput", "description": "VLLM Engine decode throughput - 2:4 Sparse (synthetic)\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax_model_len - 4096\nbenchmark_throughput {\n \"use-all-available-gpus_\": \"\",\n \"input-len\": 2,\n \"output-len\": 128,\n \"num-prompts\": 32,\n \"sparsity\": \"semi_structured_sparse_w16a16\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 889.5189608279312 tokens/s

This comment was automatically generated by workflow using github-action-benchmark.

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

smaller_is_better

Benchmark suite Current: 707295f Previous: d69a34a Ratio
{"name": "median_request_latency", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 3613.9676914999654 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 142.8064850366536 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 90.28198249961861 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 24.020879345748522 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 23.361090086372233 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2397.891458999311 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 114.26435027732684 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 78.17981800008056 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 17.43045474084683 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 15.68620587694513 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2016.9221374999324 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 79.81105100000302 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 38.60395599986077 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11.507704049974086 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11.589697739013644 ms
{"name": "median_request_latency", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11941.897211499963 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 160.31142607734 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 126.67857049927989 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 107.16286959046508 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 96.11016904795461 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11799.268781000592 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 198.17807075801892 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 150.84239999987403 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 91.49517194069021 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 93.80863680549051 ms
{"name": "median_request_latency", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5025.827374999608 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 87.85559696655885 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 60.45521899977757 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 36.04416539803316 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 30.652281846898497 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 2516.4029234997543 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 120.0964251466806 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 83.65136849943156 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 18.61376270187983 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 16.681701474682264 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5994.225282500338 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 119.5230529599985 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 79.01800599984199 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 39.19310387807475 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 38.81870025741521 ms
{"name": "median_request_latency", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 57005.43279449994 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 24093.722715536675 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 21864.01632700108 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 180.26208027034875 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 191.7838624052724 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1933.598044000064 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 79.01941300665082 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 38.0912079999689 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 10.96748650983727 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11.077961046604349 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 10730.237756500173 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 188.0490948073208 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 145.7949830000871 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 81.77788008363 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 83.61232653608496 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6329.191929499984 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 110.19734126666435 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 70.83355749995235 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 36.006174160528886 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 36.71500171271662 ms
{"name": "median_request_latency", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5176.644684499479 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 180.58307924002293 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 169.4242330004272 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 38.28789666145273 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned50\nmax-model-len - 4096\nsparsity - sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 35.51887333285672 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 6090.378854499932 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 104.95343306665745 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 66.406779999852 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 34.38797764388648 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"150,0.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 34.9214290025821 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 251562.08500000002 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 237369.02623565734 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 238758.94937499994 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 69.14240749569474 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 65.65281187879462 ms
{"name": "median_request_latency", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5639.596768999581 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 117.36010848397564 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 86.63894099936442 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 51.4501299216107 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - 2:4 Sparse\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-pruned2.4\nmax-model-len - 4096\nsparsity - semi_structured_sparse_w16a16\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 44.80312512778893 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1806.5232485000706 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 91.87236818006568 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 57.56063999979233 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 12.471127830700109 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - neuralmagic/OpenHermes-2.5-Mistral-7B-marlin\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 11.730691949105712 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 1934.132046999821 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 95.34300081003494 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 61.05815499995515 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 13.066311582103177 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - TheBloke/OpenHermes-2.5-Mistral-7B-GPTQ\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 12.350703863930315 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 7411.302156500028 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 154.31797739199678 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 128.34148150000146 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 55.501108754854705 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 55.106800078614356 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 5963.371730500057 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 127.94797432332719 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 89.11311349993412 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 40.10878089164205 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"300,1\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 40.055444239633104 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 52617.62654949996 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 32027.316234700662 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 31069.96550899987 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 93.86975104806493 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - teknium/OpenHermes-2.5-Mistral-7B\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"1500,5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 96.73101568293018 ms
{"name": "median_request_latency", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 73647.81685099979 ms
{"name": "mean_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 54364.474810331994 ms
{"name": "median_ttft_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 66604.0276350002 ms
{"name": "mean_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 67.04738576096028 ms
{"name": "median_tpot_ms", "description": "VLLM Serving - Dense\nmodel - NousResearch/Llama-2-7b-chat-hf\nmax-model-len - 4096\nsparsity - None\nbenchmark_serving {\n \"nr-qps-pair_\": \"750,2.5\",\n \"dataset\": \"sharegpt\"\n}", "gpu_description": "NVIDIA A10G x 1", "vllm_version": "0.3.0", "python_version": "3.11.4 (main, May 10 2024, 13:52:50) [GCC 9.4.0]", "torch_version": "2.3.0+cu121"} 64.26029261890845 ms

This comment was automatically generated by workflow using github-action-benchmark.

Please sign in to comment.