fix: update OpenAI model metadata for openai/gpt-oss-120b#580
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.
|
![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: fcfb46b13a
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".
| "databricks-meta-llama-3-1-405b-instruct": ["databricks"], | ||
| "databricks-meta-llama-3-1-8b-instruct": ["databricks"], | ||
| "openai/gpt-oss-120b": ["together", "groq", "baseten"], | ||
| "openai/gpt-oss-120b": ["groq","together","baseten"], |
There was a problem hiding this comment.
Changing the provider order for openai/gpt-oss-120b alters runtime behavior when API-secret lookup fails: EdgeProxyV1 falls back to endpointTypes[0] (packages/proxy/edge/index.ts lines 275-279), so this change now forces fallback traffic to Groq instead of Together. In that failure mode, requests authenticated with a Together key will be sent to Groq and fail authentication, which is a regression from the previous ordering.
Useful? React with 👍 / 👎.
Caitlin Pinn (cpinn)
force-pushed
the
chore/autofix-issue-537
branch
from
368df59 to
f8feb48
Compare
1 participant
fix: update OpenAI model metadata for openai/gpt-oss-120b
Closes #537
Source issue: #537
Summary
openai/gpt-oss-120bopenai/gpt-oss-120b2
Verified metadata
Verification notes
Verification
Sources and fields verified
input_cost_per_mil_tokens($0.15),output_cost_per_mil_tokens($0.60)input_cost_per_mil_tokens($0.15),output_cost_per_mil_tokens($0.60),input_cache_read_cost_per_mil_tokens($0.075)reasoning(true — model supports chain-of-thought)available_providers— confirms Baseten hosts the model on shared inferencesync_models cross-check
The sync_models catalog has a single entry
azure_ai/gpt-oss-120b(no Together/Groq-prefixed entry exists).input_cost_per_mil_tokensoutput_cost_per_mil_tokensinput_cache_read_cost_per_mil_tokensmax_input_tokensmax_output_tokensFields not changed
format,flavor,displayName,reasoning,max_input_tokens,max_output_tokens,available_providers,input_cache_read_cost_per_mil_tokens— all remain at current catalog values; only the two pricing fields are updated.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.