RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Kimi Dev 72B run on the 2x RTX 4090?

Yes. At 8K context the best fit is Q3_K_M, using about 38.3 GB of the 45.6 GB usable on the 2x RTX 4090. Estimated speed ceiling: ~27 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)27.3 GB Fits28.6 GB Fits36.1 GB Fits66.1 GB No37
3-bit (Q3_K_M) •37.0 GB Fits38.3 GB Fits45.8 GB No75.8 GB No27
4-bit (Q4_K_M)43.3 GB Tight44.6 GB Tight52.1 GB No82.1 GB No23
5-bit (Q5_K_M)50.2 GB No51.4 GB No58.9 GB No88.9 GB No—
6-bit (Q6_K) •61.9 GB No63.1 GB No70.6 GB No101 GB No—
8-bit (Q8_0) •73.9 GB No75.1 GB No82.6 GB No113 GB No—
16-bit (F16/BF16) •137 GB No139 GB No146 GB No176 GB No—

• real GGUF file size from unsloth/Kimi-Dev-72B-GGUF. Speed is a bandwidth ceiling (1008 GB/s), not a benchmark. Methodology.

More

Kimi Dev 72B on every GPU and Mac · Every model that runs on the 2x RTX 4090 · Try other settings in the calculator

Data: Hugging Face Hub (moonshotai/Kimi-Dev-72B), NVIDIA specs. Data as of 2026-11-17.