feat: support openai v1/completions api

[blueswhen]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @blueswhen, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly expands the API capabilities by integrating the OpenAI v1 completions API. It introduces a new endpoint, defines the necessary data models, and implements the core logic to process text completion requests, including advanced features like streaming, batching, and detailed log probability reporting. The changes also include backend enhancements to support these new functionalities efficiently and a comprehensive set of tests to ensure correctness and robustness.

Highlights

• OpenAI Completions API Endpoint: Implemented the /v1/completions endpoint, providing a standard interface for text completion requests.
• Data Model Definitions: Introduced comprehensive Pydantic models (CompletionRequest, CompletionResponse, etc.) to accurately represent the OpenAI Completions API request and response formats.
• Core Completion Logic: Developed the completions_impl function and its helpers to handle diverse completion scenarios, including support for streaming, batch processing, logprobs generation, and echo functionality.
• Backend Logprobs Support: Enhanced the request object (req.py) and model inference backend (impl_for_return_all_prompt_logprobs.py) to efficiently capture and cache prompt logprobs, crucial for accurate logprobs and echo responses.
• Extensive Test Coverage: Added a thorough suite of unit tests for the new completions API, covering various parameters like prompt types (string, token IDs, batch), stream, logprobs, echo, n, and stop_sequences.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces support for the OpenAI-compatible /v1/completions endpoint. The implementation is thorough, covering various request parameters, streaming, and batching. The code is well-structured, and the addition of extensive tests is a major strength.

I've identified one high-severity issue concerning error handling that should be addressed to ensure the API provides appropriate feedback for invalid client requests.

[gemini-code-assist]

Raising a generic Exception for a client error (e.g., invalid parameter n in stream mode) will result in a 500 Internal Server Error. This should be handled as a client-side error, returning a 400 Bad Request. The create_error_response utility function is designed for this purpose and should be used here for proper error handling and API behavior.

Suggested change

	raise Exception("stream api only support n = 1")
	return create_error_response(HTTPStatus.BAD_REQUEST, "stream api only support n = 1")