GeezerVison

Add Gemini 4 Argon to the Agents known models catalog, so it appears as a suggestion when an admin adds a chat model, and regenerate the AI Gateway price book from models.dev.

Pricing. The price book carries Google's launch rates on google, openrouter, vercel, and copilot: $2 per 1M input tokens, $10 per 1M output tokens, and $0.10 per 1M cached input tokens. Google doubles them on 2026-12-01 ($4, $20, $0.20); the weekly refresh picks that up once models.dev changes.

Output limit. overrides.jq raises the google entry from models.dev's 32,768 output tokens to 65,536, Google's documented default synchronous limit, because the limit includes thinking tokens. Its guards fail if the model or its upstream value changes. OpenRouter (65,536) and Vercel (1,000,000) keep their published values.

Thinking levels. googleSupportedThinkingLevels did not recognize the argon name token and failed closed, so chatd sent Argon no thinking_level and no include_thoughts: the native Google path dropped the reasoning effort, and neither path showed thought summaries. Argon accepts LOW, MEDIUM, and HIGH like Gemini 3.1 Pro and rejects MINIMAL, so the map now treats it like the Pro family.

Upstream drift. prices.json covers every model of its 8 providers, so the regeneration also brings the changes since the last weekly refresh, all on OpenRouter and Vercel: 14 other new entries, 26 entries models.dev delisted, and 51 price changes. The catalog picks up the same changes for 8 curated OpenRouter entries (DeepSeek, Kimi, Qwen, and GLM).

Testing

• Live requests to gemini-4-argon through the AI Gateway's Google provider (Google's OpenAI-compatible endpoint): reasoning_effort and thinking_level values low, medium, and high return 200 and minimal returns 400; include_thoughts returns thought text in streaming and non-streaming responses.
• Live max_tokens requests through the AI Gateway return 200: Google at 65,536 (streaming and non-streaming) and 98,304 (streaming, the compaction headroom); Vercel at 1,000,000, also with a 70,044-token prompt.
• Red-green: removing the argon token fails TestGoogleSupportedThinkingLevels/gemini-4-argon (expected LOW, MEDIUM, HIGH; got nil).
• Regenerating from the saved models.dev snapshot reproduces both generated files byte for byte; with the new pin, only the google Argon output limit changes and prices.json stays identical.
• go test ./scripts/aibridgepricesgen/... ./coderd/aibridge/prices/... ./coderd/x/chatd/chatprovider/... and pnpm exec vitest run --project=unit src/modules/aiModels/ pass. The pre-commit hook passed; local pre-push checks were skipped, so CI is the first full run.

Review record

• Frontend review: the only site/ change is the generated knownModelsGenerated.json, so the FE1 to FE10 checklist does not apply; no hand-written frontend code changed.
• Codex round 1 (be2c236, 3 findings): fixed the google output limit in cc0f220; declined the Vercel output limit (live requests at 1,000,000 succeed) and a curated Google reasoning effort (thought summaries already work without one, and no Google entry curates one).

Xum opened this PR on behalf of @ibetitsmike.