### The Feature

The native `minimax/` provider prefix in the cost map already includes `minimax/MiniMax-M2`, `minimax/MiniMax-M2.1`, `minimax/MiniMax-M2.5`, and `minimax/MiniMax-M3`, but there is **no entry for the released `minimax/MiniMax-M2.7`** model. Today `MiniMax-M2.7` only exists under non-native prefixes (`sambanova/MiniMax-M2.7`, `fireworks_ai/.../minimax-m2p7`, `pinstripes/ps/minimax-m2.7`), so users calling the model through the native MiniMax provider get no cost tracking or context-window metadata.

Note: closed issues #24601 and #25614 requested `openrouter/minimax/minimax-m2.7` (a different provider prefix), and #24533 is an unrelated Ollama-Cloud runtime bug — so this native `minimax/` cost-map entry is not yet tracked.

### Model details

`MiniMax-M2.7` is a text-input reasoning/thinking model with a 204,800-token context window.

**Pricing (per 1M tokens):**

``` Input: $0.30 / 1M tokens Cached Input: $0.06 / 1M tokens Output: $1.20 / 1M tokens ```

Source: https://platform.minimax.io/docs/guides/pricing-token-plan (official MiniMax platform pricing). Input/output rates ($0.30 / $1.20) are confirmed by MiniMax's official platform pricing; the cached…