Skip to content

feat: add google/gemini-3.1-flash-lite-image - #129

Merged
guillaumegay13 merged 4 commits into
mainfrom
mp/add-google-gemini-3.1-flash-lite-image
Aug 18, 2026
Merged

feat: add google/gemini-3.1-flash-lite-image#129
guillaumegay13 merged 4 commits into
mainfrom
mp/add-google-gemini-3.1-flash-lite-image

Conversation

@guillaumegay13

@guillaumegay13 guillaumegay13 commented Jul 19, 2026

Copy link
Copy Markdown
Collaborator

✨ What changed

  • Adds gemini-3.1-flash-lite-image (google) to the catalog, params cloned from google/gemini-3.5-flash.yaml.

💭 Why

The provider serves it but the catalog doesn't list it.

📝 Notes

  • Liveness ✅ only: the model answers on its real endpoint, but params are sibling-cloned, not individually probed. Review the param surface.
  • Draft PR, auto-proposed by the modelparams agent.

@guillaumegay13 guillaumegay13 added the liveness-verified Model verified live on its endpoint; params sibling-cloned, not individually probed label Jul 19, 2026
@vercel

vercel Bot commented Jul 19, 2026 •
edited
Loading

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
modelparams.dev Ready Ready Preview Aug 18, 2026 3:04pm

Request Review

@github-actions github-actions Bot added the model Add a model that's missing label Jul 19, 2026
@guillaumegay13
guillaumegay13 force-pushed the mp/add-google-gemini-3.1-flash-lite-image branch from 19a0fb5 to 31337ab Compare July 20, 2026 13:10
@guillaumegay13
guillaumegay13 force-pushed the mp/add-google-gemini-3.1-flash-lite-image branch from 31337ab to 38ac459 Compare August 17, 2026 16:15
@guillaumegay13 guillaumegay13 mentioned this pull request Aug 17, 2026
Merged
@guillaumegay13

guillaumegay13 commented Aug 17, 2026

Copy link
Copy Markdown
Collaborator Author

Probe attempt 2026-08-17: the model id is served (429 quota-exceeded, not 404), but the key's image-generation quota is exhausted so the param surface can't be verified. Holding this open rather than merging: the params here are cloned from a text model (thinkingConfig.thinkingLevel, responseMimeType, etc.) and image-generation models typically have a different surface (responseModalities, no thinking) — merging unprobed would land the wrong-surface class of bug. Re-probe when image quota resets.

@guillaumegay13
guillaumegay13 merged commit 11f6094 into main Aug 18, 2026
12 checks passed
@guillaumegay13
guillaumegay13 deleted the mp/add-google-gemini-3.1-flash-lite-image branch August 18, 2026 15:04
@guillaumegay13 guillaumegay13 mentioned this pull request Aug 18, 2026
Merged
guillaumegay13 added a commit that referenced this pull request Aug 18, 2026
@guillaumegay13 @claude
revert: remove google image-generation models from the catalog (#252)
a1b04de 
modelparams.dev tracks chat/completion params only; the discovery
pipeline's chat filter already excludes google -image models. These
three entries predate that filter (PRs #127/#128/#129, July) and were
merged today by mistake.

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

This branch was successfully deployed

1 active deployment
Preview — 62027d43 Deployed Aug 18, 2026 by vercel[bot]
View deployment
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

liveness-verified Model verified live on its endpoint; params sibling-cloned, not individually probed model Add a model that's missing

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@guillaumegay13