Skip to content

fix(gemini): update model catalog for retired/new Gemini models#11793

Merged
RomneyDa merged 3 commits intomainfrom
gemini-v1beta
Mar 25, 2026
Merged

fix(gemini): update model catalog for retired/new Gemini models#11793
RomneyDa merged 3 commits intomainfrom
gemini-v1beta

Conversation

@RomneyDa
Copy link
Collaborator

@RomneyDa RomneyDa commented Mar 25, 2026

Summary

  • Add Gemini 3.1 models (gemini-3.1-pro-preview, gemini-3.1-flash-lite-preview, gemini-3.1-flash-image-preview)
  • Replace shut-down gemini-3-pro-preview with gemini-3.1-pro-preview across codebase
  • Remove all retired Gemini 1.x models (gemini-1.0-pro, gemini-1.5-pro, gemini-1.5-flash, gemini-1.5-flash-8b) which return 404
  • Fix default model from retired gemini-pro to gemini-2.5-flash
  • Update docs, onboarding, VS Code schema, GUI model picker, VertexAI/CometAPI providers, and agent model recommendations

Fixes #10996
Fixes #10998

Context

Google retired all Gemini 1.x models and shut down gemini-3-pro-preview on March 9, 2026 (redirects to gemini-3.1-pro-preview). New 3.1 series models are now the recommended defaults. The v1beta API base was already correct.

Test plan

  • Verify Gemini provider in Add New Model UI shows updated model list
  • Verify gemini-3.1-pro-preview works with Gemini API key
  • Verify gemini-3-flash-preview works with Gemini API key
  • Verify onboarding flow with Gemini selects working models
  • Run toolSupport.test.ts tests pass

Summary by cubic

Update the Gemini model catalog to add the 3.1 series, remove retired 1.x and old 3 Pro models, and align defaults, UI, and docs with supported options. Also fixes schema and agent-model detection so users only see and select working models.

  • New Features

    • Added gemini-3.1-pro-preview, gemini-3.1-flash-lite-preview, and gemini-3.1-flash-image-preview
    • Updated model catalogs, GUI picker, VS Code schema (includes gemini-3.1-flash-image-preview), onboarding, docs (Gemini 3.1 Pro heading), and providers (Gemini, VertexAI, CometAPI)
    • Tightened agent model detection to match gemini-3.1-pro-preview and gemini-3-flash-preview only; refreshed tests and recommendations
  • Migration

    • Remove gemini-1.0-pro, gemini-1.5-pro, gemini-1.5-flash, gemini-1.5-flash-8b, and retired 2.0 image-gen variants
    • Replace any gemini-3-pro-preview with gemini-3.1-pro-preview; replace gemini-pro with gemini-2.5-flash where used

Written for commit 440ff0c. Summary will update on new commits.

Gemini 1.x models are fully retired (returning 404), gemini-3-pro-preview
was shut down March 9 2026, and new 3.1 series models were missing.

- Add gemini-3.1-pro-preview, gemini-3.1-flash-lite-preview, gemini-3.1-flash-image-preview
- Replace gemini-3-pro-preview with gemini-3.1-pro-preview everywhere
- Remove retired Gemini 1.x models (gemini-1.0-pro, gemini-1.5-pro, gemini-1.5-flash, gemini-1.5-flash-8b)
- Remove retired experimental 2.0 image gen models
- Fix default model from retired gemini-pro to gemini-2.5-flash
- Update docs, onboarding config, VSCode schema, GUI model picker, VertexAI provider
- Update isRecommendedAgentModel regex and tests

Fixes #10996, #10998
@RomneyDa RomneyDa requested a review from a team as a code owner March 25, 2026 02:32
@RomneyDa RomneyDa requested review from Patrick-Erichsen and removed request for a team March 25, 2026 02:32
@mintlify
Copy link

mintlify bot commented Mar 25, 2026

Preview deployment for your docs. Learn more about Mintlify Previews.

Project Status Preview Updated (UTC)
continue-docs 🟢 Ready View Preview Mar 25, 2026, 2:33 AM

@dosubot dosubot bot added the size:L This PR changes 100-499 lines, ignoring generated files. label Mar 25, 2026
Copy link
Contributor

@cubic-dev-ai cubic-dev-ai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

2 issues found across 15 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="extensions/vscode/config_schema.json">

<violation number="1" location="extensions/vscode/config_schema.json:973">
P2: The Gemini model enum is missing `gemini-3.1-flash-image-preview`, so a supported Gemini model is rejected by VS Code schema validation.</violation>
</file>

<file name="core/llm/toolSupport.ts">

<violation number="1" location="core/llm/toolSupport.ts:464">
P2: The new Gemini regex is too broad and will incorrectly classify `gemini-3-flash` (non-preview) as a recommended agent model.</violation>
</file>

Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.

- Add missing gemini-3.1-flash-image-preview to VS Code schema enum
- Tighten isRecommendedAgentModel regex to match 3-flash-preview (not bare 3-flash)
- Update case-insensitivity tests for new model names
- Rename test to "Gemini 3.1 Pro and 3 Flash Preview models" to match assertions
- Update chat.mdx heading from "Gemini 3 Pro" to "Gemini 3.1 Pro"
@github-project-automation github-project-automation bot moved this from Todo to In Progress in Issues and PRs Mar 25, 2026
@dosubot dosubot bot added the lgtm This PR has been approved by a maintainer label Mar 25, 2026
@RomneyDa RomneyDa merged commit 09650d4 into main Mar 25, 2026
67 of 68 checks passed
@RomneyDa RomneyDa deleted the gemini-v1beta branch March 25, 2026 03:22
@github-project-automation github-project-automation bot moved this from In Progress to Done in Issues and PRs Mar 25, 2026
@github-actions github-actions bot locked and limited conversation to collaborators Mar 25, 2026
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

lgtm This PR has been approved by a maintainer size:L This PR changes 100-499 lines, ignoring generated files.

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

Error: Gemini 3.0 Pro - Unknown error Error: Gemini 3.1 Flash (Fast) - Unknown error

2 participants