Moonshot releases Kimi K3 weights, a 2.8T MoE model with 1M-token context
HarveenChadha · x · 2026-07-27
Moonshot has released the weights and technical report for Kimi K3, its most capable model to date.
- Scale: a 2.8T MoE model with native visual understanding and a 1M-token context window.
- Architecture claim: the team says K3 delivers 2.5× more intelligence per unit of compute, not just more parameters.
- Open stack: Moonshot is also opening more of the infrastructure behind the model, including high-performance attention kernels, an MoE communication library, and tooling for running agent environments at scale.
- The post links to the model weights, tech report, and tech blog on Hugging Face and Moonshot’s site.
Related event: Moonshot releases Kimi K3 open weights amid license debate(155 posts)→
More from Infra
- 12 KV Cache Reduction Techniques Every AI Engineer Should Understand, Explained — blaizedsouza · 2026-09-11
- The shadow GPU capacity market is formalizing, with Meta selling excess compute to outside buyers — DavidLinthicum · 2026-09-11
- Engram's random reads don't suit SSDs; CPU-memory over NVLink could serve all 72 GPUs — bookwormengr · 2026-09-11
- 80% of the DIY LLM inference hype posters have already quit — it's brutally hard systems work — abhijithneil · 2026-09-11
- Hugging Face's Ultra Scale Playbook: a free book on training LLMs on GPU clusters — mdancho84 · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11