Kimi K2 Models — Hardware Requirements
8 Kimi K2 models from Moonshot AI and the community, from the smallest that runs in 1.9 GB of VRAM at Q4 up to 1026.9B parameters. Every row links to full quantization tables and GPU compatibility.
All Kimi K2 Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Kimi K2 Instruct | 1026.4B | 617.3 GB | 131K | ||
| Kimi K2 Thinking | 1026.4B | 617.3 GB | 262K | ||
| Kimi K2 Instruct 0905 | 1026.4B | 617.3 GB | 262K | ||
| Kimi K2 Base | 1026.5B | 617.3 GB | 131K | ||
| Kimi K2.5 | 1026.9B | 617.6 GB | 262K | ||
| Kimi K2.6 | 1026.9B | 617.6 GB | 262K | ||
| Kimi K2.7 Code | 1026.9B | 617.6 GB | 262K |
Community variants and fine-tunes(5)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Kimi K2.6 Eagle3 | 1.8B | 1.9 GB | 262K | ||
| Kimi K2.6 Eagle3 Mla | 3.0B | 2.6 GB | 262K | ||
| Kimi K2.6 Dspark | 3.1B | 4.1 GB | — | ||
| Kimi K2.6 DFlash | 3.5B | 2.7 GB | 262K | ||
| Kimi K2.7 Code DFlash | 3.5B | 2.7 GB | 262K |
How Kimi K2 Compares — llmrun Composite Score
Kimi K2.6 is the highest-scoring Kimi K2 model with an llmrun Composite Score of 120.9 — #15 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
Kimi K2.6120.9
Kimi K2.7 Code118.1
Kimi K2.5113.9
Kimi K2 Thinking108.6
Kimi K2 Instruct100.1
Frequently Asked Questions
- How much VRAM do I need to run a Kimi K2 model?
- The smallest Kimi K2 model, Kimi K2.6 Eagle3, runs from 1.9 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 1.1 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Kimi K2 models can I run on a 16 GB GPU?
- 5 of 12 Kimi K2 models fit in 16 GB of VRAM at Q4_K_M, including Kimi K2.6 Eagle3 Mla, Kimi K2.6 DFlash, Kimi K2.7 Code DFlash.
- What is the most popular Kimi K2 model to run locally?
- Kimi K2.5 is the most downloaded Kimi K2 model in local-friendly quantized formats. It runs from 617.6 GB of VRAM at Q4_K_M.
- How do Kimi K2 models score on benchmarks?
- Kimi K2.6 leads the family with an llmrun Composite Score of 120.9, ranking #15 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.