StepFun Step — local AI model by StepFun
Written by Jakub Rusinowski · Last updated
StepFun's open-weight mixture-of-experts models, which interleave sliding-window and full attention so the long-context cache stays small: Step 3.7 Flash (Apache-2.0) and the 600B Step 5 Preview (StepFun Model License 1.0).
Variants
The smallest StepFun Step variant needs about 112 GB of VRAM at Q4_K_S — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| Step 3.7 Flash 198B-A11B → 198B (11B active) | ~112.3 GB | 262,144 | cloud-hosted tag | |
| Step 5 Preview 600B-A27B → 600B (27B active) | ~365.6 GB | 1,048,576 | cloud-hosted tag |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run StepFun Step locally
Install Ollama, then pull the tag.
No Ollama tag. Use StepFun's official GGUFs with llama.cpp (mainline supports the Step 3.5 architecture the model uses; the card's build steps use StepFun's own fork), or vLLM / SGLang.
Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: Step 3.7 Flash 198B-A11BWeights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Applies to: Step 5 Preview 600B-A27BRecommended GPU
The cheapest catalogued GPU that runs StepFun Step locally (min 112 GB VRAM) is the NVIDIA DGX Spark (128 GB).
StepFun Step — frequently asked questions
How much VRAM does StepFun Step need?
StepFun Step needs about 112 GB VRAM at Q4_K_S quantization for its smallest variant. Variants: Step 3.7 Flash 198B-A11B (112 GB, Q4_K_S); Step 5 Preview 600B-A27B (366 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run StepFun Step on an RTX 4090 (24 GB)?
StepFun Step's smallest variant needs about 112 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for StepFun Step?
Q4_K_M is the best balance of quality and VRAM for StepFun Step in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run StepFun Step with Ollama?
StepFun Step has no local Ollama tag — the published tag is cloud-hosted, so running it sends your prompts to a hosted GPU rather than your own machine.