chore: update model catalog from bot issues#926
Conversation
github-actions
Bot
requested review from
Alex Z (CLowbrow),
aswink,
Caitlin Pinn (cpinn),
Erin McNulty (erin2722) and
Ken Jiang (knjiang)
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Codex (@codex) review |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: baabfe6082
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "Codex (@codex) review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "Codex (@codex) address that feedback".
| "multimodal": true, | ||
| "input_cost_per_mil_tokens": 0.4, | ||
| "output_cost_per_mil_tokens": 2, | ||
| "deprecation_date": "2026-08-31", |
There was a problem hiding this comment.
Mistral's legacy table has separate Deprecation and Retirement columns for these models: 2026-05-22 and 2026-08-31 respectively, and the model cards also show deprecation date 2026-05-22. Since markModelsPastDeprecationDate only marks models after deprecation_date, storing the retirement date keeps both mistral-medium-2505 and mistral-medium-2508 active in the catalog throughout the deprecation window.
Useful? React with 👍 / 👎.
#922 set deprecation_date = 2026-08-31 for mistral-medium-2505 and mistral-medium-2508, but Mistral's legacy-models table lists that as the RETIREMENT date; the deprecation date is 2026-05-22 (verified on Mistral's models overview docs). deprecation_date drives markModelsPastDeprecationDate, so the retirement value would leave these models showing active until Aug 31 even though they were deprecated in May. Use 2026-05-22 (Codex P2). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Revert #925's change to groq's $0.15/$0.60. openai/gpt-oss-120b is a single catalog entry shared by groq, together, and baseten, whose per-provider prices differ (groq 0.15/0.60, baseten 0.10/0.50). Per decision, prefer baseten pricing for this shared entry, matching main and applyBasetenPricing's behavior. This leaves the entry unchanged vs main; #925 is not applied. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
1 participant
Automated daily batch of model catalog updates from bot issues.
Included issues
Summary
mistral-medium-2505mistral-medium-2508mistral-medium-2505mistral-medium-2508groq/compoundgroq/compound-minigroq/compoundgroq/compound-mini2
openai/gpt-oss-120bopenai/gpt-oss-120b2
Verified metadata
#922: [BOT ISSUE] Mistral: add deprecation markers for mistral-medium-2505 and mistral-medium-2508
Verification notes
No LLM verification step ran; model metadata was already complete in the issue.
sync_models vs proposed update
sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.
#924: [BOT ISSUE] Groq: add missing groq/compound and groq/compound-mini models
Verification notes
No LLM verification step ran; model metadata was already complete in the issue.
sync_models vs proposed update
sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.
#925: [BOT ISSUE] Groq: openai/gpt-oss-120b has stale pricing ($0.10/$0.50 → $0.15/$0.60)
Verification notes
No LLM verification step ran; model metadata was already complete in the issue.
sync_models vs proposed update
sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.