Add as a preferred source on GoogleKimi K3.1 Pro (Max) Ranks First Among Open-Weight Agent Models
It placed 13th across all models, with results drawn from more than 17,500 extended agent-task sessions submitted by users globally.
Kimi K3.1 Pro (Max) ranked first among open-weight models on an agent benchmark, leading its open-model group in work, programming and chat and placing 13th overall.
The ranking drew on more than 17,500 extended agent-task sessions submitted by users around the world. Kimi K3.1 Pro (Max) posted +7.9% on Bash Reliability, a measure of shell-command execution, topping open models. Its controllability metric was +6.5%, also first among open models.
Moonshot AI describes Kimi K3.1 Pro as a mixture-of-experts model with 2.8 trillion total parameters and about 104 billion active in each inference. It supports a context window of about 1 million tokens and native multimodal capabilities. The company lists coding, 3D creation, financial research, factual accuracy and complex agent tasks as areas of focus.
Moonshot AI announced Thursday, Nov. 5, that it had published Kimi K3.1 Pro’s weights on Hugging Face.
Simon Yoon
ReporterSimon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.