RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Kimi VL A3B Instruct run on the RTX 4070 Super?

Yes. At 8K context the best fit is Q3_K_M, using about 8.7 GB of the 11.4 GB usable on the RTX 4070 Super. Estimated speed ceiling: ~332 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •7.0 GB Fits7.1 GB Fits7.8 GB Fits10.6 GB Tight419
3-bit (Q3_K_M) •8.6 GB Fits8.7 GB Fits9.4 GB Fits12.2 GB No332
4-bit (Q4_K_M) •10.6 GB Tight10.8 GB Tight11.5 GB No14.3 GB No262
5-bit (Q5_K_M) •12.0 GB No12.2 GB No12.9 GB No15.7 GB No—
6-bit (Q6_K) •14.1 GB No14.2 GB No15.0 GB No17.8 GB No—
8-bit (Q8_0) •16.6 GB No16.8 GB No17.5 GB No20.3 GB No—
16-bit (F16/BF16)31.4 GB No31.5 GB No32.2 GB No35.1 GB No—

• real GGUF file size from mradermacher/Kimi-VL-A3B-Instruct-GGUF. Speed is a bandwidth ceiling (504 GB/s), not a benchmark. Methodology.

More

Kimi VL A3B Instruct on every GPU and Mac · Every model that runs on the RTX 4070 Super · Other models that fit 12 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (moonshotai/Kimi-VL-A3B-Instruct), NVIDIA specs. Data as of 2026-11-17.