Skip to content

chore: add Groq Qwen3.6 preset#505

Closed
c121914yu wants to merge 1 commit into
labring:mainfrom
c121914yu:automation/plugin-model-update-20260714
Closed

chore: add Groq Qwen3.6 preset#505
c121914yu wants to merge 1 commit into
labring:mainfrom
c121914yu:automation/plugin-model-update-20260714

Conversation

@c121914yu

Copy link
Copy Markdown
Contributor

Summary

  • Add Groq qwen/qwen3.6-27b LLM preset.
  • Set the preset from Groq official model data: 131,072 context, 32,768 max output tokens, vision enabled, reasoning enabled, and tool choice enabled.

Evidence

Validation

  • node .codex/skills/model-provider-updater/scripts/model_provider_presets.mjs inventory
  • bun run test passed: 39 files / 246 tests. Existing warnings: missing optional .env.test / .env.test.local, and Vitest coverage package version mismatch.
  • bunx tsx -e "...ProviderConfigSchema.parse(groq)..." passed for the Groq provider data.
  • bun tsc --noEmit currently fails on existing sdk/factory/src/tool-factory.test.ts(285,9) stream mock typing; this is unrelated to the static Groq preset change.

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Pull request overview

Adds a new Groq LLM preset entry for qwen/qwen3.6-27b to the provider’s static model inventory so it can be selected/configured consistently alongside existing Groq presets.

Changes:

  • Add Groq preset for qwen/qwen3.6-27b with maxContext: 131072 and maxTokens: 32768.
  • Mark the model as vision: true, reasoning: true, reasoningEffort: true, and toolChoice: true, matching the Groq provider config shape used elsewhere in the file.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@c121914yu c121914yu closed this Jul 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants