Skip to content

fix(providers): correct GLM-5.2 context window to 1M - #483

Open
hongweiduan wants to merge 1 commit into
andrewyng:mainfrom
hongweiduan:fix/glm-5.2-context-window
Open

fix(providers): correct GLM-5.2 context window to 1M#483
hongweiduan wants to merge 1 commit into
andrewyng:mainfrom
hongweiduan:fix/glm-5.2-context-window

Conversation

@hongweiduan

Copy link
Copy Markdown

Summary

  • coworker/providers/matrix.py listed the OpenRouter GLM-5.2 entry with a 128K context window, but GLM-5.2 actually exposes a 1,000,000-token context.
  • Updated the entry to 1_000_000 so callers stop under-provisioning requests routed through OpenRouter.

Context

Only the OpenRouter path was affected; the native zai:glm-5.2 entry is a separate row. No behavioral change beyond the larger advertised window.

Test plan

  • python -c "from coworker.providers.matrix import MATRIX; print(MATRIX['openrouter:z-ai/glm-5.2'])"
  • Verify a >128K-token prompt routed via OpenRouter is no longer rejected by the matrix entry itself.

🤖 Generated with Claude Code

The OpenRouter GLM-5.2 entry advertised a 128K context window, but
GLM-5.2 actually supports 1,000,000 tokens. Update the matrix entry
so callers stop under-provisioning the context.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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