[Feat] Add Gemini 3.6 Flash model

[zoomote]

Opened on behalf of Naved Merchant. Follow up by mentioning @zoomote, in the web UI, or in Discord.

Related GitHub Issue

None — requested directly by a maintainer via the linked task thread (add the newly released Gemini 3.6 Flash the same way the last Gemini model was added).

Description

Adds gemini-3.6-flash to the Gemini and Vertex providers, mirroring the Gemini 3.5 Flash addition (#331) exactly: one model entry in each of packages/types/src/providers/gemini.ts and packages/types/src/providers/vertex.ts, no default-model change (the 3.5 Flash PR did not change the default either).

Specs were taken from Google's official documentation (not copied from the 3.5 entry):

• Model page (cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-6-flash): model ID gemini-3.6-flash, GA (released July 21, 2026), 1,048,576 context window, 65,536 max output tokens, text output with image/audio/video input, thinking and context caching supported.
• Thinking docs: supports thinking levels minimal/low/medium/high with a medium default — identical to 3.5 Flash.
• Pricing page (Standard PayGo): $1.50/M input, $7.50/M output, $0.15/M cached input (no >200K tier), with the same $1/M Tok/hr Flash-class cache storage convention used for cacheWritesPrice in the 3.5 entry. Note the $7.50 output price is 3.6's actual rate (3.5 Flash is $9.00).

Reviewers may want to spot-check the pricing against the Agent Platform generative-AI pricing page.

Test Procedure

• Model entries are validated by the shared types schema and provider tests: pnpm test from the repo root passes (all 10 turbo tasks; 230 types tests, 33 gemini/vertex provider tests).
• Reviewers can verify by selecting the Gemini or Vertex provider in settings and confirming Gemini 3.6 Flash appears in the model dropdown.

Pre-Submission Checklist

• Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above). (Requested directly by a maintainer; no tracking issue.)
• Scope: My changes are focused on the linked issue (one major feature/fix per PR).
• Self-Review: I have performed a thorough self-review of my code.
• Testing: New and/or updated tests have been added to cover my changes (if applicable).
• Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
• Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

No UI changes — model registry entries only. Zoo Code is a VS Code extension; the new model surfaces in the Gemini and Vertex provider model dropdowns in settings.

Documentation Updates

• No documentation updates are required. The Gemini/Vertex provider docs pages do not enumerate individual model versions ("Zoo Code supports all Gemini models... and automatically tracks Google's latest stable releases").

Additional Notes

packages/types/src/providers/vscode-llm.ts intentionally not updated: that list reflects models exposed by VS Code's built-in LM API, which does not offer Gemini 3.6 Flash yet (same reasoning as when 3.5 Flash was added).

Get in Touch

Opened via Roomote on behalf of the requester — follow up via the linked task thread in the provenance note above.

[zoomote]

No code issues found. The two new gemini-3.6-flash entries are identical between the Gemini and Vertex providers, mirror the gemini-3.5-flash structure exactly (only outputPrice differs: 7.5 vs 9), pass the shared model schema (230/230 types tests green), and require no other registration points — the model dropdowns derive from these maps. Pricing/spec numbers are externally sourced and can't be verified from the repo, so a maintainer spot-check against Google's pricing page is still worthwhile. See task

_{Reviewed e6df3ae}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!