K3 already got in the top 5 most liked models of all time on Hugging Face, just 24 hours after being released! Ahead of Llama 3, Whisper and many other great models. Incredible!

Kimi K3.1 Pro on TPUs, with day-0 support ⚡ Run it on TPU v7x with vllm-torchtpu. Great to see the continued collaboration between the @Google and @inferact teams! 🤝 Read more 👇

Kimi K3.1 Pro is now live on Alibaba Cloud Model Studio & Qwen Cloud! We're excited to partner with @Kimi_Moonshot as a Day-0 launch partner, bringing production-ready inference to developers and enterprises from day one. Build with confidence, scale with ease, and get started with 1M free tokens to explore what Kimi K3.1 Pro can do. Try it now: Model Studio: Qwen Cloud: #Kimi# #KimiK3_1Pro# #AlibabaCloud# #QwenCloud# #ModelStudio#

Kimi K3.1 Pro weights are out. New: They replaced all 24 Gated MLA layers from K3 with Gated DeepSeek Sparse Attention. The other 69 KDA layers remain unchanged. DSA uses a top-k of just 2,048 for a model with 1M context. DeepSeek is a year ahead when it comes to architecture.

Kimi k3.1 and Deepseek were distilled from GPT-6 Astra According to The Information, both Chinese AI labs used Astra's outputs to help train their models I already said this both these models have Astra taste

Kimi K3.1 Pro is among the leading models in agentic knowledge work and coding, but early signs suggest it will rank lower on AA-3D We are still collecting the preference votes for its Elo on AA-3D, our benchmark for how well AI models build interactive 3D scenes. In agentic knowledge work and coding, it is among the leading models on GDPval-AA and Terminal-Bench 4.0. Its predecessor, Kimi K3, is 12th of 22 models on AA-3D. Here is one of its AA-3D scenes: a 492 m mountain-bike descent with berms, roots, a rock garden, a drop, and a jump. A demo run rides it from top to bottom, and you can take over the steering, pedaling, and braking at any point. Congratulations to @Kimi_Moonshot on the release!