feat: sync model list — 27 models, Premium + Open Source groups#1
feat: sync model list — 27 models, Premium + Open Source groups#1Deign86 wants to merge 1 commit into
Conversation
…roups - Add taste-1 (CommandCode personal coding style model) - Add claude-opus-4.8 (complex reasoning, long-horizon agentic coding) - Add step-3.7-flash (multimodal sparse-MoE reasoning) - Update qwen-3.6-max-preview: enabled, vibe coding description - Update qwen-3.7-max: enabled, frontier coding description - Update step-3.5-flash: fast sparse-MoE description - Remove legacy claude-opus-4.5, claude-sonnet-4.5, claude-sonnet-4 - Update all model descriptions to match new categorization - Add model selector to tray app (Premium + Open Source groups) - Update credentials.json with synchronized model list - Fix config tests to match new enabled states and descriptions
|
Thank you so much for opening this PR! I’m genuinely happy to receive a contribution here — this is one of the first external PRs to the project, and I really appreciate you taking the time to inspect and improve the model list. One bit of context: CommandCode is currently updating very frequently, and this bridge is also being updated/automated quickly to stay aligned with those releases. Because of that, the baseline may have moved while you were preparing this PR, which can make contributing a bit inconvenient. Sorry about that, and thank you for working through it. I reviewed the diff, but I can’t merge this as-is because GitHub currently shows merge conflicts, and the PR also changes several things outside the intended model-list sync. Could you please adjust the PR as follows? 0. Please update the branch against the latest
|
Summary
Syncs the CommandCode bridge model catalog with the updated model list across Premium and Open Source groups.
Changes
New models added:
commandcode/taste-1— Personal coding style (CommandCode taste learning)anthropic/claude-opus-4.8— Complex reasoning, long-horizon agentic codingstepfun/Step-3.7-Flash— Multimodal sparse-MoE reasoningDescriptions updated:
Qwen/Qwen3.6-Max-Preview→ "Vibe coding, efficient agent execution" (now enabled)alibaba/qwen3.7-max→ "Frontier coding, long-horizon agent execution" (now enabled)stepfun/Step-3.5-Flash→ "Fast sparse-MoE agentic reasoning at scale"Legacy models removed:
anthropic/claude-opus-4-5-20251101(Opus 4.5)anthropic/claude-sonnet-4-5-20250929(Sonnet 4.5)anthropic/claude-sonnet-4-20250514(Sonnet 4)Tray app enhanced:
Modelssubmenu with Premium and Open Source grouped sectionsOther:
credentials.jsonsynchronized with 27 models