@nate_512The intelligence-to-cost ratio is shifting again and XIAOMI just dropped MiMo V2.6 Pro to prove it. This new flagship model scored 46 on the Artificial Analysis Intelligence Index, placing it ahead of Qwen3.8 Max and Kimi K3 while sitting just one point behind GPT-5.6 Sol. The cost profile is what makes this a practical shift for devs since each task on the Artificial Analysis Intelligence Index costs $0.13 for this model. Those performance improvements resulted from scaling a post-training self-improvement system that utilizes an Environment × Task × Grader loop applied to the pre-training foundation to push the same base further. The release features fully multimodal capabilities enabled by default and includes autonomous research capabilities designed for scientific workloads. A materials research case demonstrated the model assisting experts in designing and computationally screening new candidate materials. http://mimo.mi.com/docs/en-US/news/latest/v2-6

@nate_512The intelligence-to-cost ratio is shifting again and XIAOMI just dropped MiMo V2.6 Pro to prove it. This new flagship model scored 46 on the Artificial Analysis Intelligence Index, placing it ahead of Qwen3.8 Max and Kimi K3 while sitting just one point behind GPT-5.6 Sol. The cost profile is what makes this a practical shift for devs since each task on the Artificial Analysis Intelligence Index costs $0.13 for this model. Those performance improvements resulted from scaling a post-training self-improvement system that utilizes an Environment × Task × Grader loop applied to the pre-training foundation to push the same base further. The release features fully multimodal capabilities enabled by default and includes autonomous research capabilities designed for scientific workloads. A materials research case demonstrated the model assisting experts in designing and computationally screening new candidate materials. http://mimo.mi.com/docs/en-US/news/latest/v2-6