Kimi K3.1 Pro

Kimi K3.1 Pro is an open language model from Moonshot AI and the first model of the K3.1 series. It was post-trained from Kimi K3 with large-scale reinforcement learning, and the authors call it their most capable model of the 3-trillion-parameter class. It has about 2.8 trillion parameters, about 104 billion of them active at each step, and a context window of one million tokens. It reads text and images and answers in text. The authors name coding, 3D creation, financial research, factual accuracy and agentic tasks as the areas of improvement. The weights are stored in a compact MXFP4 format.

Reasoning depth is set per request: low, high or maximum, and maximum is the default. In multi-turn chats and tool calls, the full reply from the model, including its reasoning and tool calls, has to be passed back unchanged. The authors recommend running it with vLLM, SGLang or TokenSpeed. The card does not state hardware requirements; with this size it is a server-level model. The model is also available through the Moonshot AI API, which is compatible with the OpenAI and Anthropic formats.

The weights are public without any acceptance step. The Kimi K3.1 Pro License lets you use, change, fine-tune and sell the model for free, with conditions: keep the license notice; a business that gives others API-style access to the model and has revenue above the license threshold over 12 months needs a separate agreement with Moonshot AI; and a commercial product with more than 100 million monthly users or very large monthly revenue must show the name "Kimi K3.1-Pro" in its interface. Internal use and use through Moonshot AI's own products or certified partners are exempt from the last two rules. Because of these conditions we mark commercial use as limited. We found no territorial restrictions in the license text; read the license before use.