Moonshot launches Kimi K3 with 2.8T parameters, 1M context, and native multimodality
mervenoyann · x · 2026-07-27
Moonshot AI has released Kimi K3, an open-weight native multimodal model with 2.8T parameters, 104B active parameters, and a 1M-token context window.
The model supports text, images, and video natively, uses MXFP4 weights via QAT, and activates 16 of 896 experts per token. Moonshot says Kimi K3 is its most capable model yet and is designed for long-horizon coding, knowledge work, and reasoning under the modified Kimi K3 license.
Related event: Moonshot releases Kimi K3 open weights amid license debate(155 posts)→
More from coding & agent
- Delegating scaling-law asymptote fitting to Codex while planning a birthday party — dlwh · 2026-10-27
- Fern 0.1: a from-scratch pretrained model with up to 10M token context — vjkaruna · 2026-10-27
- Celesto launches CI runners nearly 90% cheaper than GitHub Actions, with agent sandboxes — aniketmaurya · 2026-10-27
- Runway launches App Studio: build no-code creative apps that generate image, video and audio — runwayml · 2026-10-27
- Supabase founders on rebuilding the backend for a billion AI-built apps — Latent Space · 2026-10-27
- Dev scans 5,000 MCP servers daily: 10% dead, 30% auth-gated, then ClaudeBot shows up — art2meta · 2026-10-27