Kimi-K3.1-Pro is now open-weights
Kimi has made its Kimi-K3.1-Pro model available with open weights, marking the first release in the Kimi-K3.1 series. The model is a 3-trillion-parameter, natively multimodal agentic system with a 1-million-token context window. It claims improvements in coding, 3D creation, financial research, and factual accuracy, along with enhanced agentic capabilities. Kimi-K3.1-Pro uses Gated DeepSeek…
Key points
- Uses Gated DeepSeek Sparse Attention (Gated DSA) for efficient long-context processing, replacing full-attention layers from Kimi-K3
- Available via Kimi’s API with OpenAI/Anthropic compatibility; benchmark results show strong performance but rely on internal evaluations
The model is now accessible via the Kimi platform API, with OpenAI/Anthropic-compatible endpoints. Benchmark results show strong performance across coding, science, and multimodal tasks, though comparisons rely on internal evaluations and third-party harnesses. Kimi-K3.1-Pro is recommended for use with specific inference engines and supports structured outputs, tool calls, and context caching. The model is released under the Kimi-K3.1-Pro License, and downloads have reached 797 in the past month.
Kimi-K3.1-Pro is now open-weights
huggingface.co · 4 November 2026
Loading the full article…
This text was published by huggingface.co. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Ricursive Intelligence releases Ripple-65B Preview, a 65B model with 1M context · 1 src
- Pangram launches AI image and video detectors with 99.96% accuracy · 2 src
- Cognition says SWE-2 Flash runs three times faster and costs one-fifth · 1 src
- A new way of measuring AI models suggests China's threat to OpenAI and Anthropic may have been overblown · 4 src
- structRFM model trains on 21 million RNA pairs · 1 src
Comments
via GitHub Discussions