Changes / Feeds / MiniMaxAI/MiniMax-M2.7
MiniMaxAI/MiniMax-M2.7
3 changes to a tracked key since 2026-04-09, newest first. Head as last read: d494266a4affc0d2995ba1fa35c8481cbd84294b.
GPU memory to fine-tune it, card by card: MiniMax-M2.7 on /m/.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2026-04-20 | x_NXMC-2026-0035 breaking | context window 196608 to 204800 tokens (2026-04-20).d7153aa1bfdc -> d494266a4aff |
| 2026-04-09 | x_NXMC-2026-0034 breaking | the chat template changed and renders differently (multi-turn, tool-call); prompts rendered with the old one no longer match what the model now expects, so cached prompts and eval sets need re-rendering (2026-04-09).4c9387fe99ab -> d7153aa1bfdc |
| 2026-04-09 | x_NXMC-2026-0033 | declared dtype unset to "bfloat16" (2026-04-09).a0ca8c65a932 -> 4c9387fe99ab |
Pin it, and let a bot move the pin
from_pretrained("MiniMaxAI/MiniMax-M2.7", revision="d494266a4affc0d2995ba1fa35c8481cbd84294b")With the Renovate preset, a pull request moves this pin when the head moves, and links back here.
[](https://www.neurarch.com/changes-feeds/models/MiniMaxAI/MiniMax-M2.7.html)