diff --git a/.buildkite/test-pipeline.yaml b/.buildkite/test-pipeline.yaml
index 9991060a31621..aa2509f22f0cd 100644
--- a/.buildkite/test-pipeline.yaml
+++ b/.buildkite/test-pipeline.yaml
@@ -92,7 +92,9 @@ steps:
   - VLLM_ATTENTION_BACKEND=FLASH_ATTN pytest -v -s basic_correctness/test_chunked_prefill.py
 
 - label: Core Test # 10min
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
+  amd_gpus: 4   # Just for the sake of queue testing
   fast_check: true
   source_file_dependencies:
   - vllm/core
@@ -105,6 +107,7 @@ steps:
   working_dir: "/vllm-workspace/tests"
   fast_check: true
   mirror_hardwares: [amd]
+  amd_gpus: 2   # Just for the sake of queue testing
   source_file_dependencies:
   - vllm/
   - tests/entrypoints/llm
@@ -186,6 +189,7 @@ steps:
   - pytest -v -s engine test_sequence.py test_config.py test_logger.py
   # OOM in the CI unless we run this separately
   - pytest -v -s tokenization
+  working_dir: "/vllm-workspace/tests" # optional
 
 - label: V1 Test
   #mirror_hardwares: [amd]
@@ -230,6 +234,7 @@ steps:
     - python3 offline_inference/profiling.py --model facebook/opt-125m run_num_steps --num-steps 2
 
 - label: Prefix Caching Test # 9min
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
   source_file_dependencies:
   - vllm/
@@ -248,6 +253,7 @@ steps:
     - VLLM_USE_FLASHINFER_SAMPLER=1 pytest -v -s samplers
 
 - label: LogitsProcessor Test # 5min
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
   source_file_dependencies:
   - vllm/model_executor/layers
@@ -269,7 +275,9 @@ steps:
     - pytest -v -s spec_decode/e2e/test_eagle_correctness.py
 
 - label: LoRA Test %N # 15min each
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
+  amd_gpus: 8
   source_file_dependencies:
   - vllm/lora
   - tests/lora
@@ -295,7 +303,9 @@ steps:
   - pytest -v -s compile/test_full_graph.py
 
 - label: Kernels Test %N # 1h each
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
+  amd_gpus: 8
   source_file_dependencies:
   - csrc/
   - vllm/attention
@@ -305,6 +315,7 @@ steps:
   parallelism: 4
 
 - label: Tensorizer Test # 11min
+  working_dir: "/vllm-workspace/tests"
   mirror_hardwares: [amd]
   soft_fail: true
   source_file_dependencies:
@@ -355,6 +366,7 @@ steps:
     - pytest -v -s encoder_decoder
 
 - label: OpenAI-Compatible Tool Use # 20 min
+  working_dir: "/vllm-workspace/tests" 
   fast_check: false
   mirror_hardwares: [ amd ]
   source_file_dependencies:
diff --git a/.buildkite/test-template.j2 b/.buildkite/test-template.j2
index ce448836a8278..573675d67f86b 100644
--- a/.buildkite/test-template.j2
+++ b/.buildkite/test-template.j2
@@ -7,7 +7,7 @@ steps:
   - label: ":docker: build image"
     depends_on: ~
     commands:
-      - "docker build --build-arg max_jobs=16 --tag {{ docker_image_amd }} -f Dockerfile.rocm --progress plain ."
+      - "docker build --build-arg max_jobs=16 --tag {{ docker_image_amd }} -f Dockerfile.rocm  --target test --progress plain ."
       - "docker push {{ docker_image_amd }}"
     key: "amd-build"
     env: