fix api_cli & qwen25 parser

[shihaobai]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @shihaobai, 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 focuses on improving the tool call parsing infrastructure within the lightllm server. It introduces support for additional tool call parser models via the command-line interface, updates the API response schema to use a more standardized term for tool call completion reasons, and refines the parsing mechanism for Qwen25 models to ensure more accurate and robust tool call detection.

Highlights

• API CLI Enhancements: Expanded the command-line interface to support new tool call parser types, specifically adding 'deepseekv3' and 'qwen' as valid choices for the --tool_call_parser argument.
• API Model Update: Updated the ChatCompletionResponseChoice model by changing the finish_reason literal from 'function_call' to 'tool_calls' for better consistency and clarity.
• Qwen25 Parser Refinement: Adjusted the Qwen25ToolCallParser by removing newline characters from bot_token and eot_token definitions, and commenting out the tool_call_separator, to refine its parsing logic.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

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 fixes and enhancements for tool call parsing. It adds support for deepseekv3 and qwen tool call parsers. It also updates the API models to use tool_calls as a finish_reason, aligning with current standards. The parsing logic for qwen25 has been improved for robustness. My review includes a suggestion to improve code clarity in the Qwen25Detector by adding a comment explaining a workaround, which will aid future maintenance.

[gemini-code-assist]

While commenting out self.tool_call_separator works, it makes the code's intent less clear because it relies on a side-effect of the base class's implementation (treating \n as normal text). For better maintainability, it would be clearer to add a comment explaining why this line is commented out.

Suggested change

	# self.tool_call_separator = "\n"
	# The separator for Qwen2.5 tool calls is a newline. This is commented out to allow
	# the base streaming parser to handle the newline as normal text. The subsequent
	# `<tool_call>` token will then correctly trigger a new tool call parse. This is a
	# workaround for the base parser's logic with this specific format.
	# self.tool_call_separator = "\n"