chore: sync new models#1009
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
![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: 2b7463d50f
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".
| "input_cost_per_mil_tokens": 1.4, | ||
| "output_cost_per_mil_tokens": 4.4, | ||
| "input_cache_read_cost_per_mil_tokens": 0.26, | ||
| "input_cache_read_cost_per_mil_tokens": 0.14, |
There was a problem hiding this comment.
For cached-input requests routed to the non-Fireworks zai-org/GLM-5.2 entry, this value should not be copied from Fireworks: accounts/fireworks/models/glm-5p2 already carries the $0.14 rate, while the mapped providers for this id are Baseten/Together and both publish cached input at $0.26/M (Baseten, Together). Lowering this shared entry to 0.14 undercounts cached-prefix cost by about 46%; please leave 0.26 here unless the providers on this id change.
Useful? React with 👍 / 👎.
2b7463d to
d9dbbef
Compare
1 participant
Automated sync of model metadata and newly available models from the approved provider allowlist.