docs: add MiniMax-M3 and update minimax default model

[octo-patch]

Summary

• Add MiniMax-M3 (latest flagship) to the minimax and minimax-cn provider docs and use it as the default in the quickstart code samples.
• Drop the older MiniMax-M2.5 / M2.5-highspeed / M2.1 / M2 rows from the model tables since they are no longer recommended for new integrations.
• Keep MiniMax-M2.7 and MiniMax-M2.7-highspeed as previous-generation options.
• Update the provider unit tests to use MiniMax-M3 as the default model and assert that M3, M2.7, and M2.7-highspeed all resolve through @ai-sdk/openai-compatible.

The auto-generated model-provider-registry.generated.ts and model-provider-types.generated.ts are intentionally left alone — the existing EXTRA_PROVIDER_REGISTRY overrides take care of the runtime registration regardless of which models models.dev currently lists, so this PR only touches the user-facing docs and tests.

Test plan

• pnpm vitest run src/registries/model-provider-registry-minimax.spec.ts passes (8/8) inside packages/core.
• npx biome check clean on the modified spec file.
• npx prettier --check clean on the modified markdown after applying prettier --write.

---

Summary by cubic

Make MiniMax-M3 the default model for minimax and minimax-cn, updating quickstart samples and provider tests; remove M2.5/M2.1/M2 from docs while keeping M2.7 and M2.7-highspeed as previous-gen options. Tests now verify M3, M2.7, and M2.7-highspeed resolve via @ai-sdk/openai-compatible; no runtime registry changes.

^{Written for commit 5845c67. Summary will update on new commits.}

Summary by CodeRabbit

• New Features

  • Added support for the new MiniMax-M3 model for the MiniMax provider.
  • Updated the default MiniMax model from M2.7 to M3.

• Documentation

  • Updated MiniMax and MiniMax (China) provider documentation pages with the latest supported models list and refreshed default configuration examples. Older model variants (M2.5, M2.1, M2) are no longer supported.

[changeset-bot]

🦋 Changeset detected

Latest commit: 5845c67

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@voltagent/core	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: b0d51614-3bee-447e-8265-57bc51815e65

📥 Commits

Reviewing files that changed from the base of the PR and between 74fe5e9 and 5845c67.

📒 Files selected for processing (4)

• .changeset/MiniMax-m3-update.md
• packages/core/src/registries/model-provider-registry-minimax.spec.ts
• website/models-docs/providers/minimax-cn.md
• website/models-docs/providers/minimax.md

---

📝 Walkthrough

Walkthrough

This PR updates the MiniMax provider to use MiniMax-M3 as the default model instead of MiniMax-M2.7. The change is reflected across registry tests, provider documentation for both standard and China variants, and a corresponding changeset entry documenting the patch release.

Changes

MiniMax-M3 Default Model Update

Layer / File(s)	Summary
Release notes for MiniMax-M3 addition .changeset/MiniMax-m3-update.md	Changeset entry documents the @voltagent/core patch release for MiniMax-M3 addition and default model switch.
Registry test validation for MiniMax-M3 default packages/core/src/registries/model-provider-registry-minimax.spec.ts	MiniMax provider registry tests are updated to validate MiniMax-M3 resolution across basic loading, China provider, MINIMAX_BASE_URL override, missing API key failure, and adapter selection scenarios. Supported model list is reduced to MiniMax-M3, MiniMax-M2.7, and MiniMax-M2.7-highspeed variants.
Provider documentation updates for MiniMax-M3 website/models-docs/providers/minimax-cn.md, website/models-docs/providers/minimax.md	Both main and China provider documentation are updated to use minimax/MiniMax-M3 and minimax-cn/MiniMax-M3 in quick-start examples and models tables, removing older model variants and listing only MiniMax-M3, MiniMax-M2.7, and MiniMax-M2.7-highspeed.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Possibly related PRs

• VoltAgent/voltagent#1167: Updates MiniMax provider tests and documentation to expect MiniMax-M3 as the default model, directly related to the MiniMax registry model resolution validation framework.

Poem

🐰 A model update hops along the way,
M3 takes the lead, hooray, hooray!
Tests and docs dance in perfect time,
The MiniMax default now prime! ✨

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The pull request title directly summarizes the main changes: adding MiniMax-M3 support and updating the default model for minimax providers.
Description check	✅ Passed	The pull request description provides comprehensive information including summary of changes, test plan results, and aligns well with the repository's template structure.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[cubic-dev-ai]

No issues found across 4 files

_{Re-trigger cubic}