Can Step 3.7 Flash run on the RTX 4060 Ti 16GB?
Not fully in VRAM. Even at 2-bit with 8K context, Step 3.7 Flash needs about 74.5 GB, more than the 15.2 GB usable on the RTX 4060 Ti 16GB. It can still run with offload (slower): Q4_K_M with part of the model in system RAM, about 0.2–0.4 tokens/sec (estimate). Details and command.
Runs with offload (slower)
Runs with offload 4-bit (Q4_K_M) at 8K context: 5 of 45 layers on the RTX 4060 Ti 16GB (about 13.8 GB) and the other 40 in system RAM (about 105 GB). Needs a PC with at least 128 GB of system RAM. Estimated speed: 0.2–0.4 tokens/sec (very slow: fine for batch jobs, painful for chat), versus a bandwidth ceiling of ~2 tok/s if it fit fully in VRAM.
llama-server -hf stepfun-ai/Step-3.7-Flash-GGUF:Q4_K_M -c 8192 -ngl 5 --no-mmproj
Estimate, not a benchmark. The command downloads the GGUF from Hugging Face. The speed range spans dual-channel DDR4-3200 to DDR5-6000 system RAM; your CPU, RAM speed and llama.cpp version move it. If llama.cpp runs out of GPU memory, lower -ngl by 2-3; if VRAM is left over, raise it for more speed. Recent llama.cpp builds can also pick the split automatically. How we estimate offload.
+ 3.7 GB with image input. The figures on this page are for text only. With image input enabled, the vision projector (mmproj) file loads alongside the model. llama.cpp's -hf downloads and loads it automatically unless you add --no-mmproj. Calculator with image input on.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | 73.0 GB No | 74.5 GB No | 83.5 GB No | 120 GB No | — |
| 3-bit (Q3_K_M) | 94.4 GB No | 95.9 GB No | 105 GB No | 141 GB No | — |
| 4-bit (Q4_K_M) | 117 GB No | 119 GB No | 128 GB No | 164 GB No | — |
| 5-bit (Q5_K_M) | 136 GB No | 138 GB No | 147 GB No | 183 GB No | — |
| 6-bit (Q6_K) | 156 GB No | 158 GB No | 167 GB No | 203 GB No | — |
| 8-bit (Q8_0) | 202 GB No | 203 GB No | 212 GB No | 248 GB No | — |
| 16-bit (F16/BF16) | 378 GB No | 379 GB No | 388 GB No | 424 GB No | — |
• real GGUF file size. Speed is a bandwidth ceiling (288 GB/s), not a benchmark. Methodology.
Hardware that does fit it
A100 80GBH100 PCIeH100 SXMRTX PRO 6000 BlackwellDGX SparkRyzen AI Max+ 395 128GB
What fits instead
- Cheapest listed GPU that runs Step 3.7 Flash: DGX Spark (128 GB, launch MSRP $3,999) at Q3_K_M, fully in VRAM at 8K context.
- Largest popular model that fits the RTX 4060 Ti 16GB: gpt-oss 20B (21B, MoE) at Q8_0, about 12.3 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
Step 3.7 Flash on every GPU and Mac · Every model that runs on the RTX 4060 Ti 16GB · Other models that fit 16 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (stepfun-ai/Step-3.7-Flash), NVIDIA specs. Model data last checked 2026-11-18; refreshed hourly.