feat(providers): add MiniMax M2.7 models

[penso]

Summary

• Add MiniMax-M2.7 and MiniMax-M2.7-highspeed to the static model catalog
• Add missing MiniMax-M2.1-highspeed variant
• Model IDs confirmed against MiniMax API docs

Validation

Completed

• cargo check -p moltis-providers passes
• Model IDs verified against MiniMax API documentation

Remaining

• cargo +nightly-2025-11-30 fmt --all -- --check
• cargo +nightly-2025-11-30 clippy -Z unstable-options --workspace --all-features --all-targets --timings -- -D warnings
• cargo test

Manual QA

1. Configure MiniMax API key in moltis.toml
2. Open model picker — verify M2.7 and M2.7 Highspeed appear at top of MiniMax list
3. Send a message using MiniMax-M2.7 — verify completion works

🤖 Generated with Claude Code

[greptile-apps]

Greptile Summary

This PR adds three new entries to the MINIMAX_MODELS static catalog in crates/providers/src/lib.rs: MiniMax-M2.7, MiniMax-M2.7-highspeed, and the previously-missing MiniMax-M2.1-highspeed. The change is minimal, follows the established pattern, and the existing starts_with("MiniMax-") context-window branch will automatically assign the correct 204_800 token limit to the new M2.7 models.

• New models are inserted at the correct position (newest version first, highspeed variant immediately after its base model), maintaining the consistent descending-version ordering already used by the existing list.
• The context-window comment on line 652 (// MiniMax M2/M2.1/M2.5: 204,800.) is now incomplete — it does not mention M2.7 even though the branch it annotates covers all MiniMax- prefixed IDs. The logic is correct; only the comment needs updating.
• Three validation items remain unchecked in the PR description (cargo fmt, cargo clippy, cargo test). Per project guidelines in CLAUDE.md, all three must pass before merging.

Confidence Score: 4/5

• Safe to merge after confirming the remaining validation checks (fmt, clippy, tests) pass — the code change itself is correct and low-risk.
• The change is a pure additive data-only edit to a static array. No logic, no control flow, and no API surface is modified. The only concern is the unchecked validation checklist items in the PR description — the project's CLAUDE.md mandates fmt, clippy, and test gates before merging, and those are still marked as remaining.
• No files require special attention beyond the minor comment staleness on line 652 of crates/providers/src/lib.rs.

Important Files Changed

Filename	Overview
crates/providers/src/lib.rs	Adds MiniMax-M2.7, MiniMax-M2.7-highspeed, and the missing MiniMax-M2.1-highspeed entries to MINIMAX_MODELS. Ordering is correct (newest first, highspeed immediately after base). The existing starts_with("MiniMax-") context-window branch handles M2.7 automatically, but the accompanying comment is now stale.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[Model ID lookup] --> B{starts_with MiniMax-?}
    B -- Yes --> C[Return 204_800 context window]
    B -- No --> D[Other provider checks...]

    subgraph MINIMAX_MODELS catalog
        M27["MiniMax-M2.7 ✨ new"]
        M27H["MiniMax-M2.7-highspeed ✨ new"]
        M25["MiniMax-M2.5"]
        M25H["MiniMax-M2.5-highspeed"]
        M21["MiniMax-M2.1"]
        M21H["MiniMax-M2.1-highspeed ✨ new"]
        M2["MiniMax-M2"]
    end

    MINIMAX_MODELS --> B

Loading

Comments Outside Diff (1)

1. crates/providers/src/lib.rs, line 652 (link)

   Stale context-window comment after adding M2.7

   The comment on this line lists only M2/M2.1/M2.5 but M2.7 models added in this PR now also resolve to 204_800 via the same starts_with("MiniMax-") branch. The logic is correct, but the comment is misleading — a future reader might wonder whether M2.7 has a different context window and why it isn't listed here.

_{Reviews (1): Last reviewed commit: "feat(providers): add MiniMax M2.7 and mi..." | Re-trigger Greptile}

[codspeed-hq]

Merging this PR will not alter performance

✅ 39 untouched benchmarks
⏩ 5 skipped benchmarks¹

---

_{Comparing glimmer-variraptor (b47776a) with main (919c443)²}

Footnotes

1. 5 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on main (b47776a) during the generation of this report, so 919c443 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!