moonshotai/Kimi-K3.1-Pro
main boardmodel
Kimi K3.1 Pro is an open-weight, natively multimodal agentic model, and our most capable 3T-class model to date. Mixture-of-experts, 2780B total (896 experts, 16 per token). Context: 1M tokens (config). License: kimi-k3.1-pro.
- Org
- moonshotai
- Released
- 2026-11-02
- Status
- main board
- License
- kimi-k3.1-pro (restricted)
- Parameters
- 2780.1B
- Architecture
- MoE · 896 experts, 16/token · KimiK31ForConditionalGeneration
- Context
- 1,048,576 tokens (config.json)
- Task
- image-text-to-text · vision
- Signal
- 2089 HF likes · 46766 downloads · 1140 X likes · 3 X post(s)
Runs on
estimate from params; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~1670.0 GB (estimated size): no listed device
- Q8 ~2955.8 GB (estimated size): no listed device
- FP16 ~5562.1 GB (estimated size): no listed device
Links
First published here 2026-11-02 · updated 2026-11-06 · markdown · JSON