Can I run Kimi Dev 72B locally?
Kimi Dev 72B by moonshotai has 73B parameters. At 4-bit with an 8K context it needs about 44.6 GB of GPU memory. It fits on 25 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/Kimi-Dev-72B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 27.3 GB | 28.6 GB | 36.1 GB | 66.1 GB |
| 3-bit (Q3_K_M) • | usable | 37.0 GB | 38.3 GB | 45.8 GB | 75.8 GB |
| 4-bit (Q4_K_M) | good, most popular | 43.3 GB | 44.6 GB | 52.1 GB | 82.1 GB |
| 5-bit (Q5_K_M) | very good | 50.2 GB | 51.4 GB | 58.9 GB | 88.9 GB |
| 6-bit (Q6_K) • | near lossless | 61.9 GB | 63.1 GB | 70.6 GB | 101 GB |
| 8-bit (Q8_0) • | effectively lossless | 73.9 GB | 75.1 GB | 82.6 GB | 113 GB |
| 16-bit (F16/BF16) • | original weights | 137 GB | 139 GB | 146 GB | 176 GB |
Which GPUs and Macs run Kimi Dev 72B?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Questions
How much VRAM does Kimi Dev 72B need?
About 44.6 GB at 4-bit (Q4_K_M) with an 8K context: 41.4 GB of weights (estimated), 2.5 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 75.1 GB.
What is the smallest GPU that runs Kimi Dev 72B?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the H100 PCIe (80 GB).
Can I run Kimi Dev 72B on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M2 Max 96GB or larger, using the default macOS GPU memory limit.
Model facts
| Hugging Face repo | moonshotai/Kimi-Dev-72B |
| Parameters | 72,706,203,648 (73B) |
| Layers / hidden size | 80 / 8192 |
| Attention | GQA; 64 heads, 8 KV heads, head dim 128 |
| KV cache per token (FP16) | 320.0 KiB |
| Max context | 131,072 tokens |
| License | mit |
| Released on HF | Jun 16, 2025 |
| Downloads (30d) / likes | 3,425 / 394 |
Sources: Hugging Face model API · config.json (moonshotai/Kimi-Dev-72B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.