Add FastAPI v1/completions/ endpoint

[athitten]

Important

The Update branch button must only be pressed in very rare occassions.
An outdated branch is never blocking the merge of a PR.
Please reach out to the automation team before pressing that button.

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Collection: [Note which collection this PR will affect]

Changelog

• Add specific line by line info of high level changes in this PR.

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this 

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

• New Feature
• Bugfix
• Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

• Related to # (issue)

[Glorf]

should be logprobs: int = 1

[Glorf]

should be zipped with choices and for each choice we should have:
{"logprobs": {"token_logprobs": [array_of_token_logprobs]}}

To fix the problem, we need to remove the unused import statement. This will clean up the code and remove the unnecessary dependency, making the code easier to read and maintain.

• Locate the import statement from pathlib import Path on line 12.
• Remove this line from the file.

To fix the problem, we need to ensure that the jinja2 environment used to render the template has autoescape enabled. This can be done by using the select_autoescape function when creating the Template object. This function will automatically enable escaping for HTML and XML files, which are the most common targets for XSS attacks.

We will modify the code to create a jinja2 environment with autoescape enabled and then use this environment to get the template and render it. This change will be made in the apply_chat_template function.

[marta-sd]

Maybe we can set the default to 0? I'm not sure if this is in the official API spec, but lm-eval-harness assumes that 0 is the default and 1 needs to be passed (here)

Edit: I checked and it actually should be null by default (see here).

[agronskiy]

@marta-sd I looked at it, it seems to me that the call stack of it will go to pytriton.ModelClient.infer_batch instead of pytriton.AsyncioModelClient and will block. See https://github.com/NVIDIA/NeMo/pull/12101/files#diff-f70646f35e4a50b01c01caf162262447c66f8f54e3b1a582e9da8ff080fc5b48R128-R129. ModelClient.infer_batch is synchronous operation.

[agronskiy]

nit: this might get blocking too, it's recommended to use aihttp instead of requests inside async functions.

[github-actions]

This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days.