Skip to content

docs: list feature-gated models#182

Open
peter941221 wants to merge 2 commits into
xiaguan:mainfrom
peter941221:docs/list-supported-model-lines
Open

docs: list feature-gated models#182
peter941221 wants to merge 2 commits into
xiaguan:mainfrom
peter941221:docs/list-supported-model-lines

Conversation

@peter941221
Copy link
Copy Markdown

The README supported-models table drifted behind the server's actual model detection logic.

Failure mechanism:

  • pegainfer-server/src/server_engine.rs already exposes feature-gated DeepSeekV2Lite and KimiK2 model lines
  • the top-level README only listed Qwen3, Qwen3.5, and DeepSeek-V4-Flash
  • readers could miss that the server can also detect DeepSeek-V2-Lite and Kimi-K2 configs when built with the matching feature flags

Semantic change:

  • add DeepSeek-V2-Lite and Kimi-K2 to the README supported-models table
  • clarify that feature-gated model lines require rebuilding pegainfer-server with the matching --features ... flag before launch

Testing:

  • source verification against pegainfer-server/src/server_engine.rs:10-18, :44-58, and :75-92

Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

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

Code Review

This pull request updates the README.md file to document support for DeepSeek-V2-Lite and Kimi-K2 models. Feedback was provided regarding an inconsistency in the supported models table, where the string "feature-gated" was incorrectly placed in the Params column instead of the Status column. A suggestion was provided to align these rows with the correct table structure and include the actual parameter counts.

Comment thread README.md Outdated
@peter941221 peter941221 marked this pull request as ready for review May 29, 2026 10:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant