Moonshot AI · 1T (32B active) · Mixture of Experts
Coding-focused agentic MoE built on K2.6 — 32B active, long-horizon coding with ~30% fewer thinking tokens Check if your GPU or Mac can run Kimi K2.7 Code locally — 573.9 GB min, 956.5 GB recommended.
2026-06256K context
Mixture of Experts
Total experts: 384
Active experts: 8
Active params: 32.0B
| Quant | Bits | VRAM | Quality | Status |
|---|---|---|---|---|
| Q2_K | 2 | 329.3 GB | low | — |
| Q3_K_M | 3 | 460.8 GB | moderate | — |
| Q4_K_M | 4 | 526.6 GB | good | — |
| Q5_K_M | 5 | 658.1 GB | good | — |
| Q6_K | 6 | 789.6 GB | excellent | — |
| Q8_0 | 8 | 1052.6 GB | excellent | — |
| F16 | 16 | 2104.7 GB | lossless | — |
Can I run Kimi K2.7 Code locally?
- Can I run Kimi K2.7 Code locally?
- Kimi K2.7 Code needs about 573.9 GB of memory at a minimum and 956.5 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does Kimi K2.7 Code need?
- At Q4_K_M, Kimi K2.7 Code uses about 526.6 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.