Can Step3 VL 10B run on the RTX 3090 Ti?
Yes. At 8K context the best fit is Q8_0, using about 9.9 GB of the 22.8 GB usable on the RTX 3090 Ti. Estimated speed ceiling: ~116 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|
| 2-bit (Q2_K) | 4.8 GB Fits | 5.4 GB Fits | 8.7 GB Fits | 264 |
| 3-bit (Q3_K_M) • | 5.1 GB Fits | 5.7 GB Fits | 9.0 GB Fits | 244 |
| 4-bit (Q4_K_M) • | 5.9 GB Fits | 6.5 GB Fits | 9.9 GB Fits | 200 |
| 5-bit (Q5_K_M) • | 6.7 GB Fits | 7.3 GB Fits | 10.6 GB Fits | 172 |
| 6-bit (Q6_K) • | 7.5 GB Fits | 8.1 GB Fits | 11.5 GB Fits | 150 |
| 8-bit (Q8_0) • | 9.4 GB Fits | 9.9 GB Fits | 13.3 GB Fits | 116 |
| 16-bit (F16/BF16) • | 31.8 GB No | 32.3 GB No | 35.7 GB No | — |
• real GGUF file size from seanbailey518/Step3-VL-10B-GGUF. Speed is a bandwidth ceiling (1008 GB/s), not a benchmark. Methodology.
More
Step3 VL 10B on every GPU and Mac · Every model that runs on the RTX 3090 Ti · Other models that fit 24 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (stepfun-ai/Step3-VL-10B), NVIDIA specs. Data as of 2026-11-17.