MiniMax publishes MiniMax-M3 model on Hugging Face
MiniMax published the MiniMax-M3 model on Hugging Face under the MiniMaxAI organisation with an image-text-to-text pipeline tag and the transformers library. Tags describe it as multimodal, MoE, agent-capable, with coding, video, and conversational use, and reference an associated paper.
PUBLISHED2026-06-02
OBSERVED2026-08-11
AGE2mo
SOURCES1
- Pipeline tag: image-text-to-text
- Library: transformers
- Tags: multimodal, moe, agent, coding, video, conversational, custom_code
- Tags include eval-results and paper reference arxiv:2606.13392
COMMUNITY
No curated reactions recorded for this event. Facts and takes are kept in separate layers — community context is added by hand, never blended into the record above.