Moonshot Releases Kimi K3 Model Weights on Hugging Face
Hugging Face ● Covered by 33 sources
Moonshot released Kimi K3, a 2.8-trillion-parameter open-weight multimodal model with a 1-million-token context window, available on Hugging Face. The model uses a Mixture-of-Experts architecture that activates 16 out of 896 experts, achieving approximately 2.5 times better scaling efficiency than its predecessor Kimi K2. The open release enables researchers and developers to build and deploy frontier AI systems for coding, knowledge work, and reasoning tasks.
Why it matters
Moonshot released Kimi K3 weights on Hugging Face, giving developers direct access to its Chinese open-weight model as the open-weight policy debate intensifies.
Related stories
Moonshot's Kimi K3 Open Model Released
Kimi · 2 months ago ·
7
Moonshot AI Releases Kimi K3: A 2.8 Trillion Parameter Open MoE Model With Kimi Delta Attention and 1M Context
MarkTechPost · 2 months ago ·
9
Kimi K3, and what we can still learn from the pelican benchmark
Simon Willison's Weblog · 2 months ago ·
37