Moonshot's Kimi K3 repository goes live with a 2.8T-parameter model card
tokenbender · x · 2026-07-27
Moonshot AI’s Kimi K3 repository is now live, with the model card showing it as a 2.8T-parameter open-weight multimodal agentic model.
- The screenshot highlights Kimi K3 as Moonshot’s most capable model so far.
- It is described as built on Kimi Delta Attention and Attention Residuals, with native vision support and a 1-million-token context window.
- The Hugging Face page also shows links to the model, homepage, and community resources, signaling the launch of the public repo and documentation.
Related event: Moonshot AI Releases Kimi-K3 Open Weights on Hugging Face(11 posts)→
More from Models
- Kimi K3 reportedly improves training efficiency by 2.5× — zephyr_z9 · 2026-07-27
- Kimi K3 goes live on Baseten Model APIs at day 0 — baseten · 2026-07-27
- NVIDIA distills Cosmos3 Super image-to-video to 4 steps with a 64B model — multimodalart · 2026-07-27
- Kimi K3 goes live on Modal with custom DFlash speculative decoding for lossless speedup — AAAzzam · 2026-07-27
- Kimi K3 with 2.8T parameters and 1M context now supported on vLLM — ricklamers · 2026-07-27
- Kimi K3 could become a cheap distillation base for personal, local models — victormustar · 2026-07-27