Moonshot makes Kimi K3 officially open weights with a 1M-token context window
scaling01 · x · 2026-07-27
Moonshot’s Kimi K3 is now officially open weights.
The attached model card screenshot describes Kimi K3 as a native multimodal agentic model with about 2.8T parameters, built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes). The card also claims a 1-million-token context window and positions it as the company’s most capable model so far, aimed at long-horizon coding, knowledge work, and reasoning.
The post itself is brief, but the screenshot makes clear that the key news is the release of the weights and the model’s frontier-style positioning.
Related event: Moonshot releases Kimi K3 open weights amid license debate(155 posts)→
More from Models
- MissionReport benchmark: only 3 of 540 AI-generated government reports pass all criteria — Exp_Mark · 2026-10-27
- Local beats cloud: Qwen 3.8 FN handles private student data as well as frontier models — dr_lm · 2026-10-27
- Aidan Gomez: model's CoT said no tool calls right before making them — aidangomez · 2026-10-27
- Nat Lambert: Cohere's model is narrow RL training on internal infra ill-suited for open weights — natolambert · 2026-10-27
- OpenAI teases DOTS is coming to Europe 'soon' — mark_k · 2026-10-27
- Clifford Pickover challenges AI to decipher a message he encoded in Norse runes — pickover · 2026-10-27