MiniMax M3.1 Flash lands on Runware with 1M context, multimodal input
MiniMax said its M3.1 Flash model is now available on inference platform Runware, accessible via the model string "minimax:m3.1@flash", thanking Runware for shipping it quickly. In the same post, MiniMax said the model accepts text, image and video input, supports a 1M-token context window with up to 524K output tokens, and offers adjustable reasoning depth. The vendor also cited a score of 42 on the Artificial Analysis Intelligence Index, 72.5% on AutomationBench-AA and 73.7% on DeepSWE, and described the model as cheaper per token than MiniMax; these figures come from MiniMax's own post and have not been independently confirmed.
The listing gives developers a concrete hosted endpoint for a long-context, multimodal MiniMax model, though its performance and pricing claims currently rest on the vendor's own statement.
One model string away: "minimax:m3.1@flash" Thanks @runware for shipping M3.1 Flash so fast 🚀
Runware: MiniMax M3.1 Flash from @MiniMax_AI is now on Runware 🧠 - Text, image & video input - 1M context, up to 524K output tokens - Adjustable reasoning depth - 42 on the Artificial Analysis Intelligence Index - 72.5% AutomationBench-AA, 73.7% DeepSWE - Cheaper per token than MiniMax