All models › moonshot
Arena score1475#25 of 416 in Chat, paid models included
Size2779.9 B105.8 B active per token (mixture of experts)
4-bit download1685 GBweights only, Q4_K_M
Max context1024Ktokens, from the model config
Where to use it for free
- The lab's own chat: www.kimi.com (free with daily limits)
- Download and run it at home: official weights · GGUF files for Ollama and LM Studio
Arena scores
Rank among all models, paid ones included. Source: Arena leaderboard dataset.
Speed on popular hardware
| Hardware | GPU memory | Fits? | Speed |
|---|---|---|---|
| RTX 3060 12 GB | 12 GB | Does not fit | — |
| RTX 4060 Ti 16 GB | 16 GB | Does not fit | — |
| RTX 3090 | 24 GB | Does not fit | — |
| RTX 4090 | 24 GB | Does not fit | — |
| RTX 5090 | 32 GB | Does not fit | — |
| RX 7900 XTX | 24 GB | Does not fit | — |
| 2 × RTX 3090 | 48 GB | Does not fit | — |
| 4 × RTX 3090 | 96 GB | Does not fit | — |
| Mac M4 Max 128 GB | 96 GB | Does not fit | — |
| Ryzen AI Max+ 395 128 GB | 96 GB | Does not fit | — |
| 8 × H100 SXM | 640 GB | Does not fit | — |
Estimate at 4-bit with 32K context, 64 GB of DDR5 for offload. Speed ≈ memory bandwidth × 70% ÷ data read per token; real numbers vary by about ±30%.
GPU memory needed
| Level | 8K context | 32K context | 128K context | Quality vs full |
|---|---|---|---|---|
| Q3_K_M | 1391 GB | 1394 GB | 1405 GB | +4.7% noticeable loss |
| Q4_K_M | 1686 GB | 1689 GB | 1699 GB | +1.2% very small loss |
| Q5_K_M | 1978 GB | 1981 GB | 1991 GB | +0.4% near-lossless |
| Q6_K | 2281 GB | 2284 GB | 2294 GB | +0.2% near-lossless |
| Q8_0 | 2955 GB | 2958 GB | 2968 GB | ≈ 0% same as full |
| F16 | 5561 GB | 5564 GB | 5574 GB | ≈ 0% same as full |
Weights + context cache (from the model's architecture) + runtime. Quality: increase in perplexity against the full model, from llama.cpp tests, scaled by model size. Simulate your own setup.
Coding score over time
160017001800
MayJunJulAugSeptOct
Kimi K3 (Max) 1655Best paid model
Arena score from 20 Apr 2026 to 18 Oct 2026.
Similar free models in Chat
- GLM-5.3 (Max) 1471 · 753.3 B
- GLM-5.3-Flash 1471 · 321.3 B
- GLM-5.2 (Max) 1470 · 753.3 B
- MiMo-V2.5-Pro 1465 · 1023.2 B
- deepseek-v4.1-flash-max 1463 · 763.2 B
- GLM-5.1 1462 · 753.9 B