Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
There was a problem hiding this comment.
Pull request overview
This PR updates the project’s MiniMax defaults and supporting documentation/tests to use MiniMax-M3 instead of MiniMax-M2.7, aligning the configured default model strings and examples across the codebase.
Changes:
- Switch MiniMax default model strings from
MiniMax-M2.7toMiniMax-M3in provider config and the MiniMax adapter. - Update tests that assert MiniMax default model behavior.
- Refresh environment examples/docs and add a design note describing the default switch.
Reviewed changes
Copilot reviewed 10 out of 10 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| README.md | Updates .env.local guidance and recommended setup to reference MiniMax-M3. |
| lib/ai-providers/provider-config.ts | Updates MiniMax provider default model to MiniMax-M3. |
| lib/ai-providers/minimax-adapter.ts | Updates the MiniMax adapter’s DEFAULT_MODEL to MiniMax-M3. |
| lib/tests/validation.test.ts | Updates validation default-model assertion to MiniMax-M3. |
| lib/tests/provider-config.test.ts | Updates provider-config default-model assertions to MiniMax-M3. |
| lib/tests/minimax-adapter.test.ts | Updates mocked MiniMax responses/expectations to MiniMax-M3. |
| docs/superpowers/specs/2026-06-04-minimax-m3-default-design.md | Adds a design note documenting the default change to M3. |
| CLAUDE.md | Updates environment variable docs to reference MiniMax-M3. |
| .env.local.example | Updates example default model and client hint to MiniMax-M3. |
| .env.example | Updates example default model and client hint to MiniMax-M3. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| | `AI_PROVIDER` | recommended | `minimax`, `grok`, or `gemini`; determines which server-side text provider adapter is used | | ||
| | `NEXT_PUBLIC_AI_PROVIDER` | recommended | Set this to match `AI_PROVIDER` for consistent client/server provider behavior in Phase 1 | | ||
| | `AI_DEFAULT_MODEL` | recommended | Override provider default model (for Phase 1, `MiniMax-M2.7`) | | ||
| | `AI_DEFAULT_MODEL` | recommended | Override provider default model (currently `MiniMax-M3`) | |
| <sup>\**</sup> `GEMINI_API_KEY` is still required if image generation should work, because `app/api/generate-image/route.ts` remains Gemini-backed. | ||
|
|
||
| > Recommended Phase 1 setup: `AI_PROVIDER=minimax`, `NEXT_PUBLIC_AI_PROVIDER=minimax`, `AI_DEFAULT_MODEL=MiniMax-M2.7`, `MINIMAX_API_KEY=...`. Keep `GEMINI_API_KEY` set if you want image generation, and optionally keep `XAI_API_KEY` available for Grok fallback/testing. | ||
| > Recommended setup: `AI_PROVIDER=minimax`, `NEXT_PUBLIC_AI_PROVIDER=minimax`, `AI_DEFAULT_MODEL=MiniMax-M3`, `MINIMAX_API_KEY=...`. Keep `GEMINI_API_KEY` set if you want image generation, and optionally keep `XAI_API_KEY` available for Grok fallback/testing. |
| - `AI_PROVIDER`: Server-side text provider selection (`minimax`, `grok`, or `gemini`) | ||
| - `NEXT_PUBLIC_AI_PROVIDER`: Set this to match `AI_PROVIDER` for consistent client/server provider behavior in Phase 1 | ||
| - `AI_DEFAULT_MODEL`: Optional text model override (Phase 1 default: `MiniMax-M2.7`) | ||
| - `AI_DEFAULT_MODEL`: Optional text model override (current default: `MiniMax-M3`) |
|
|
||
| ## Context | ||
|
|
||
| LongCut currently defaults MiniMax text generation to `MiniMax-M2.7`. MiniMax has released `MiniMax-M3`, and the project should use M3 by default while preserving existing environment-variable overrides for rollback or experimentation. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
MiniMax-M2.7toMiniMax-M3.Official MiniMax Docs Check
MiniMax-M3as a supported model with a 1,000,000-token context window and show the OpenAI-compatible base URL ashttps://api.minimax.io/v1.MiniMax-M3as a validmodelvalue and usehttps://api.minimax.io/v1/chat/completions.model="MiniMax-M3"and documentreasoning_splitas a supported M3 request parameter.max_completion_tokensfor new integrations, but still documentmax_tokensas a legacy generation-length parameter. This PR intentionally preserves the existing request payload shape and only changes the default model.References:
Test Plan
node --import tsx --test lib/__tests__/minimax-adapter.test.ts lib/__tests__/provider-config.test.ts lib/__tests__/validation.test.tsnpm run lint(0 errors, 5 existing warnings)