Skip to content

fix: use configurable Groq model (llama-3.3-70b-versatile deprecated) - #14

Merged
kx7m2qd merged 1 commit into
mainfrom
fix/groq-model-config
Sep 3, 2026
Merged

fix: use configurable Groq model (llama-3.3-70b-versatile deprecated)#14
kx7m2qd merged 1 commit into
mainfrom
fix/groq-model-config

Conversation

@kx7m2qd

@kx7m2qd kx7m2qd commented Sep 3, 2026

Copy link
Copy Markdown
Owner

What

  • Groq chat model is now process.env.GROQ_MODEL || "openai/gpt-oss-120b" instead of the three hardcoded llama-3.3-70b-versatile references.
  • GROQ_MODEL documented in .env.example.

Why

Groq deprecated the llama-3.3-70b-versatile chat model — every AI analysis call was failing with model_not_found even though the API key was valid. Verified live that openai/gpt-oss-120b works with the app's exact request shape (JSON response mode, system prompt, tools-capable).

Verification

  • Live Groq call with the new model + JSON mode: 200 OK, valid JSON returned
  • npm test — all pass
  • npm run lint — clean
  • npm run build — clean

The hardcoded llama-3.3-70b-versatile chat model no longer exists on
Groq, so every AI analysis call failed with model_not_found. Default to
openai/gpt-oss-120b (verified working with JSON mode) and allow
overriding via GROQ_MODEL.
@kx7m2qd
kx7m2qd merged commit 9821165 into main Sep 3, 2026
1 check passed
@kx7m2qd
kx7m2qd deleted the fix/groq-model-config branch September 3, 2026 17:21
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