Moonshot open-weights Kimi K3.1 Pro: 2.8T-param MoE with 1M context
scaling01 · x · 2026-11-04
Moonshot AI has released open weights for Kimi K3.1 Pro on Hugging Face, its most capable 3T-class natively multimodal agentic model. Key specs: MoE with 2.8T total / 104B activated parameters, 896 experts (16 per token), 1M-token context, KDA + Gated DSA attention replacing Gated MLA, MoonViT-V2 vision encoder, and MXFP4/MXFP8 quantization-aware training. Post-trained from Kimi K3 with large-scale joint RL across hundreds of downstream domains, with claimed gains in coding, 3D creation, financial research, factual accuracy, and agentic capabilities.
Related event: Moonshot AI Open-Sources Kimi K3.1 Pro: 2.8T-Parameter MoE with 1M Context(15 posts)→
More from Models
- Tencent's Hy4 adds vision and computer use — but long-task reliability is the open question — Traditional_Air319 · 2026-11-05
- Nathan Lambert launches free open model economy dashboard with paid analysis series — natolambert · 2026-11-05
- Underdog alert: Xiaomi MiMo keeps competing with top models on cost-adjusted benchmarks — VictorSanh · 2026-11-05
- Pangram Image 1.0 detects 99.2% of AI images at 0.27% false positive rate — TuhinChakr · 2026-11-05
- Tencent reportedly rolls out AI model trained with limited cybersecurity data amid hacking fears — pstAsiatech · 2026-11-05
- Users can't tell frontier models from 50x cheaper ones, orchestration may win — RihardJarc · 2026-11-05