B.AI Adds Another Free Model, Xiaomi MiMo-V2.5 Fully Multimodal Flagship Model Now Open to All

Odaily News: The B.AI platform has announced that Xiaomi's open-source flagship multimodal model, MiMo-V2.5, is now officially available for free, full-scale access. As a 310B sparse MoE natively multimodal large model, MiMo-V2.5 supports text, image, video, and audio inputs across all modalities, and features a 1M ultra-long context window with rapid response capabilities. It is deeply optimized for high-intensity real-world scenarios such as multimodal agents, intelligent audio-video analysis, and general-purpose programming.

Starting today, developers can unlock the cutting-edge power of Xiaomi's flagship multimodal model at zero cost through the official API suite on the B.AI platform, integrating top-tier multimodal capabilities into their own applications and workflows with no barriers. Log in to chat.b.ai/chat now to experience the new all-modal intelligent experience brought by Xiaomi MiMo-V2.5 for free.