Fireworks AI and Modular deployed day-zero hosting support.
MiniMax M3 is live on Fireworks. Day-0, fastest endpoint for the MiniMax series.
→ Top open-weight model on the Artificial Analysis index → 512K context, native image + video input → MSA sparse attention: 9× faster prefill, 15× faster decode → Priced at parity with M2.7
Long-horizon agents, full-repo understanding, multimodal coding all in one model.
Learn more: https://fireworks.ai/models/fireworks/minimax-m3
M3 from @MiniMax_AI is now live on Modular Cloud, day zero. Open weights, 1M-token context, multimodal. MSA is a new attn architecture and getting its performance benefits requires whole-stack optimization.
That's what we built Modular for. Thrilled to be a launch partner!👇
M3 open weight just dropped and it's live on @Modular cloud on day zero
with up to a 1M-context and MSA architecture kernel-to-cloud optimization is exactly what M3 needs
glad to have @Modular with us from the start
@clattner_llvm @MiniMax_AI open weights is cool but that 1M context is gonna eat RAM for breakfast
@clattner_llvm @MiniMax_AI 1M context is crazy, but whole-stack just for attention?
imagine explaining this to someone still on GPT-4
@FireworksAI_HQ kimi when
No Digg Deeper questions have been answered for this story yet.
