feat(providers): refresh Qwen, Cohere, and Mistral catalogs#102489
Conversation
openclaw-barnacle
Bot
added
docs
|
Codex review: needs maintainer review before merge. Reviewed July 9, 2026, 3:15 AM ET / 07:15 UTC. Summary PR surface: Source +93, Tests +143, Docs +8. Total +244 across 20 files. Reproducibility: not applicable. as a bug reproduction, but the catalog gap is source-reproducible: current main lacks the new Qwen/Cohere IDs and still carries stale Mistral Small metadata. Review metrics: 1 noteworthy metric.
Root-cause cluster Members:
Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything. Merge readiness Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch. Rank-up moves:
Risk before merge
Maintainer options:
Next step before merge
Security Review detailsBest possible solution: Land the provider-owned catalog, docs, and focused test refresh after exact-head CI completes, while preserving existing defaults and letting the linked issue close through the merged PR. Do we have a high-confidence way to reproduce the issue? Not applicable as a bug reproduction, but the catalog gap is source-reproducible: current main lacks the new Qwen/Cohere IDs and still carries stale Mistral Small metadata. Is this the best way to solve the issue? Yes. Updating the provider-owned static catalogs, docs, and focused tests is the maintainable fix; live-only discovery would add provider lifecycle complexity for a curated catalog refresh. AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against 363cf0461f25. Label changesLabel changes:
Label justifications:
Evidence reviewedPR surface: Source +93, Tests +143, Docs +8. Total +244 across 20 files. View PR surface stats
What I checked:
Likely related people:
What the crustacean ranks mean
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics. How this review workflow works
|
![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: 1a42e91e52
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| name: QWEN_37_MAX_MODEL_ID, | ||
| reasoning: true, | ||
| input: ["text"], | ||
| cost: QWEN_DEFAULT_COST, |
There was a problem hiding this comment.
When users select qwen/qwen3.7-max on the Standard endpoint, this new Standard-only catalog row reuses the zero-cost placeholder, and calculateCost multiplies model.cost.* directly, so usage reporting will show every run as $0 even though Qwen lists nonzero pay-as-you-go pricing for the model ($2.50/M input, $7.50/M output: https://docs.qwencloud.com/developer-guides/getting-started/pricing). Please give this Standard-only model its actual cost (or endpoint-specific pricing) so cost displays are not misleading.
Useful? React with 👍 / 👎.
clawsweeper
Bot
added
proof: sufficient
|
Merged via squash.
|
…#102489) * feat(providers): refresh popular model catalogs * chore: leave release changelog to release automation * docs: refresh provider docs map
1 participant
Closes #102460
What Problem This Solves
Qwen and Cohere users cannot select several current models from OpenClaw's direct-provider catalogs, current Qwen Coding Plan models are hidden by stale endpoint filtering, and Mistral Small 4 is advertised with outdated context and pricing metadata.
Why This Change Was Made
Refresh the provider-owned static catalogs from current official contracts: add Qwen 3.7 Max/Plus with endpoint-aware availability, add Cohere North Mini Code 1.0 with its binary reasoning mapping, and align both Mistral Small 4 IDs with current context, pricing, and reasoning behavior. Existing provider defaults, auth flows, protocols, and config surfaces stay unchanged.
User Impact
Users can select Qwen 3.7 Plus on Coding Plan or Standard endpoints, Qwen 3.7 Max on Standard endpoints, and Cohere North Mini Code directly. Mistral users get accurate Small 4 limits/pricing plus the pinned
mistral-small-2603model with the same default-off reasoning controls asmistral-small-latest.Evidence
qwen3.7-maxandqwen3.7-plus: HTTP 200, expectedOC-QWEN-LIVEcompletion.mistral-small-latestandmistral-small-2603: HTTP 200, expectedOC-MISTRAL-LIVEcompletion.max_tokensandreasoning_effort(none/high).tbx_01kx2rcatyw1a4wfpbkpgpr4ez:OPENCLAW_CHECK_CHANGED_REMOTE_CHILD=1 OPENCLAW_CHANGED_LANES_RAW_SYNC=1 corepack pnpm check:changed: passed, including extension production/test typechecks, extension lint, import-cycle, database-first, and repository guards.