Moonshot releases Kimi K3 with 2.8T MoE, 1M context and 2.5× efficiency gain
joeddav · x · 2026-07-28
Moonshot releases Kimi K3 weights and technical report.
- Kimi K3 is described as the company’s most capable model: a 2.8T MoE with native visual understanding and a 1M-token context window.
- The new architecture claims 2.5× better intelligence per unit of compute than Kimi K2, not just a bigger model.
- Moonshot is also opening more of the stack behind K3, including high-performance attention kernels, an MoE communication library, and infrastructure for running agent environments at scale.
Related event: Moonshot AI Releases Open-Weight Kimi K3 Model(36 posts)→
More from Models
- A repost claims Anthropic’s Opus 5 regresses badly despite benchmark gains — rickasaurus · 2026-07-28
- Polymarket prices a 76% chance Moonshot ships another Kimi K model by September — Polymarket · 2026-07-28
- Moonshot releases open weights for 2.8T-parameter Kimi K3 — Polymarket · 2026-07-28
- Kimi K3 reveals its pre-training mix across text, code, math, knowledge, and vision — stochasticchasm · 2026-07-28
- The Verge says Moonshot’s open Kimi K3 could undercut closed U.S. AI models — The Verge AI · 2026-07-28
- Kimi K3 lands on Fireworks AI for inference and training — omarsar0 · 2026-07-28