Can Kimi VL A3B Thinking run on the RTX 3090?
Yes. At 8K context the best fit is Q8_0, using about 17.2 GB of the 22.8 GB usable on the RTX 3090. Estimated speed ceiling: ~294 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | 6.6 GB Fits | 6.7 GB Fits | 7.4 GB Fits | 10.2 GB Fits | 832 |
| 3-bit (Q3_K_M) | 8.3 GB Fits | 8.4 GB Fits | 9.1 GB Fits | 12.0 GB Fits | 638 |
| 4-bit (Q4_K_M) | 10.2 GB Fits | 10.3 GB Fits | 11.0 GB Fits | 13.8 GB Fits | 510 |
| 5-bit (Q5_K_M) | 11.7 GB Fits | 11.8 GB Fits | 12.5 GB Fits | 15.4 GB Fits | 438 |
| 6-bit (Q6_K) | 13.4 GB Fits | 13.5 GB Fits | 14.2 GB Fits | 17.0 GB Fits | 380 |
| 8-bit (Q8_0) | 17.1 GB Fits | 17.2 GB Fits | 17.9 GB Fits | 20.7 GB Tight | 294 |
| 16-bit (F16/BF16) | 31.4 GB No | 31.5 GB No | 32.2 GB No | 35.1 GB No | — |
• real GGUF file size. Speed is a bandwidth ceiling (936 GB/s), not a benchmark. Methodology.
More
Kimi VL A3B Thinking on every GPU and Mac · Every model that runs on the RTX 3090 · Other models that fit 24 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (moonshotai/Kimi-VL-A3B-Thinking), NVIDIA specs. Data as of 2026-11-17.