MiniMax M3.1 Flash is out, and we've open-sourced a DFlash 2 drafter for it under Apache 2.0. It makes generation up to 3.5× faster, and it's lossless. Congrats to @MiniMax_AI on the release!
Inco AI open-sources a DFlash 2 drafter for MiniMax M3.1 Flash
AIInco AI says it has open-sourced a DFlash 2 drafter for MiniMax M3.1 Flash under Apache 2.0. The company claims the drafter makes generation up to 3.5 times faster while remaining lossless.
AIWhy it matters.
The post names the drafter's license and the speedup and losslessness claims, which help readers judge whether it fits their own MiniMax M3.1 Flash deployment.
Inco AIVerified on X
@inco_ai
MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token. Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82.2 on CyberGym. Efficient by design: MiniMax Sparse Attention and DSpark speculative decoding improve inference efficiency. Hugging Face: https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash ModelScope: https://modelscope.cn/models/MiniMax/MiniMax-M3.1-Flash API: http://platform.minimax.io M Plan: https://minimax.io/m-planMiniMax MCode for desktop: https://agent.minimax.io/downloadView quoted post on X
Source details
Inco AI · x.comPublished · added here
Original title: MiniMax M3.1 Flash is out, and we've open-sourced a DFlash 2 drafter for it under Apache 2.0. It makes generation up to 3.5× faster, and ...