Request: add Command Code and OpenStarry providers

Request: add Command Code and OpenStarry providers

Hi maintainers! I'd like to request adding two OpenAI-compatible gateway providers to models.dev. I'm not deeply familiar with the TOML schema edge cases, so I'm handing this over for you to review and add properly.

Provider 1: Command Code

• Name: Command Code
• API: https://api.commandcode.ai/provider/v1
• Env: COMMAND_CODE_API_KEY
• npm: @ai-sdk/openai-compatible
• Docs: https://commandcode.ai/docs/reference/cli/models
• Model list endpoint: GET /models — public, no key required
• Reasoning: reasoning_effort request body field, values low | medium | high
• Models (~58): Claude (sonnet-5/4-6, fable-5, opus-5/4-8/4-7, haiku-4-5), GPT (5.6-sol/terra/luna, 5.5, 5.4, 5.3-codex, 5.4-mini), DeepSeek (v4-pro, v4-flash, v4-flash-vision-exp), Kimi (K3, K2.7-Code/-Highspeed, K2.6, K2.5), GLM (5.3, 5.2, 5.2-Fast, 5.1, 5), MiniMax (M3, M2.7, M2.5), MiMo (v2.5-pro, v2.5), Qwen (3.8-Max, 3.8-27B, 3.7-Max/Plus/Flash, 3.6-Max-Preview, 3.6-Plus), Step (3.7/3.5-Flash), Gemini (3.7/3.6/3.5/3.5-lite/3.1-lite), plus stealth/ox-alpha (free), sakana/fugu-ultra, nvidia/nemotron-3-ultra, thinkingmachines/inkling(-small), poolside/laguna-s-2.1-free, meta/muse-spark-1.x, xai/grok-4.5/4.6, tencent/hy3-paid

Provider 2: OpenStarry

• Name: OpenStarry
• API: https://api.openstarry.com/v1
• Env: OPENSTARRY_API_KEY
• npm: @ai-sdk/openai-compatible
• Docs: https://api.openstarry.com/openstarry-docs.html
• Model list endpoint: GET /models — requires API key
• Models (10): deepseek-v4-flash(-0731), deepseek-v4-pro-0813, glm-5.3, kimi-k2.6, kimi-k2.7-code, mimo-v2.5, qwen3.7-plus, MiniMax-M2.7(-highspeed)

Draft PR for reference

I prepared PR #5347 (kept as draft) containing:

• Sync provider modules: packages/core/src/sync/providers/commandcode.ts and openstarry.ts
• Generated catalogs via bun models:sync: 58 model TOMLs under providers/commandcode/models/, 10 under providers/openstarry/models/
• New base metadata: models/stealth/ox-alpha.toml
• Both providers registered in packages/core/src/sync/index.ts

Feel free to adopt, regenerate from the endpoints above, or discard as you see fit.

Validation note

CI validate currently fails on that draft with:
providers/commandcode/models/sakana/fugu-ultra.toml → limit.output Required

Root cause appears to be an upstream data gap: models/sakana/fugu-ultra.toml on dev has [limit] context = 1_000_000 but no output, and my generated file references it via base_model. It seems no other provider currently references it as a base model, so the gap was never triggered before. Might be worth adding an output value there regardless.

Thanks!

Review the draft changes in packages/core/src/sync/providers/commandcode.ts, packages/core/src/sync/providers/openstarry.ts, packages/core/src/sync/index.ts, and the generated provider model TOMLs. Run bun models:sync and CI validate, starting with the reported sakana/fugu-ultra.toml limit.output failure. Done means both providers and their catalogs validate successfully and are registered in the sync index.