Changes / Feeds / MiniMaxAI/MiniMax-M2
MiniMaxAI/MiniMax-M2
6 changes to a tracked key since 2025-10-23, newest first. Head as last read: 757303d492a50514c312788b5247a4f696a4c6a3.
GPU memory to fine-tune it, card by card: MiniMax-M2 on /m/.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2025-12-08 | x_NXMC-2025-0049 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 (2025-12-08).9354a5cc95c7 -> f62f643a718e |
| 2025-11-07 | x_NXMC-2025-0048 | the chat template text changed but rendered byte-identically on the fixed conversation set (system+user, multi-turn, tool call) (2025-11-07).78ba357af9df -> 9354a5cc95c7 |
| 2025-11-06 | x_NXMC-2025-0047 breaking | EOS declared as 200020 (was unset); generation that relied on the library default now stops on this token; BOS declared as 200019 (was unset); model_type "minimax" to "minimax_m2"; a different class is instantiated from the same id (2025-11-06).44cefa66f81e -> 78ba357af9df |
| 2025-10-27 | x_NXMC-2025-0046 breaking | model_type "mixtral" to "minimax"; a different class is instantiated from the same id (2025-10-27).e9774173c852 -> 44cefa66f81e |
| 2025-10-27 | x_NXMC-2025-0045 | generation defaults changed (top_k unset to 40); a call that does not pass these explicitly now generates differently (2025-10-27).cab79a3e1dc0 -> e9774173c852 |
| 2025-10-25 | x_NXMC-2025-0044 breaking | architectures ["MiniMaxM2MiniForCausalLM"] to ["MiniMaxM2ForCausalLM"]; a different class is instantiated from the same id (2025-10-25).49303277636d -> da5eb8f2eb03 |
Pin it, and let a bot move the pin
from_pretrained("MiniMaxAI/MiniMax-M2", revision="757303d492a50514c312788b5247a4f696a4c6a3")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.html)