Can I run Step 3.7 Flash locally?
Step 3.7 Flash by stepfun-ai has 201B parameters. At 4-bit with an 8K context it needs about 119 GB of GPU memory. It fits on 12 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from stepfun-ai/Step-3.7-Flash-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 73.0 GB | 74.5 GB | 83.5 GB | 120 GB |
| 3-bit (Q3_K_M) • | usable | 90.1 GB | 91.6 GB | 101 GB | 137 GB |
| 4-bit (Q4_K_M) | good, most popular | 117 GB | 119 GB | 128 GB | 164 GB |
| 5-bit (Q5_K_M) | very good | 136 GB | 138 GB | 147 GB | 183 GB |
| 6-bit (Q6_K) | near lossless | 156 GB | 158 GB | 167 GB | 203 GB |
| 8-bit (Q8_0) • | effectively lossless | 201 GB | 203 GB | 212 GB | 248 GB |
| 16-bit (F16/BF16) • | original weights | 376 GB | 378 GB | 387 GB | 423 GB |
Which GPUs and Macs run Step 3.7 Flash?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Prefer an API?
StepFun: Step 3.7 Flash is on OpenRouter at $0.20 per 1M input tokens and $1.15 per 1M output tokens. Price details and history →
Questions
How much VRAM does Step 3.7 Flash need?
About 119 GB at 4-bit (Q4_K_M) with an 8K context: 115 GB of weights (estimated), 3.6 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 203 GB.
What is the smallest GPU that runs Step 3.7 Flash?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the H200 (141 GB).
Can I run Step 3.7 Flash on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M2 Ultra 192GB or larger, using the default macOS GPU memory limit.
Model facts
| Hugging Face repo | stepfun-ai/Step-3.7-Flash |
| Parameters | 201,365,316,160 (201B) |
| Layers / hidden size | 45 / 4096 |
| Attention | MHA + sliding window; 64 heads, 64 KV heads, head dim 128; sliding window 512 |
| KV cache per token (FP16) | 1536.0 KiB (sliding layers capped at the window) |
| Max context | 262,144 tokens |
| License | apache-2.0 |
| Released on HF | May 23, 2026 |
| Downloads (30d) / likes | 9,644 / 463 |
Sources: Hugging Face model API · config.json (stepfun-ai/Step-3.7-Flash) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.