Evidence: benchmark results vendor-reported
Step 3.7 Flash 198B-A11B — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
A 196B-parameter MoE language model with a 1.8B vision encoder (198B together) and about 11B active per token: 288 routed experts with 8 active plus a shared expert, 45 layers of which 12 use full attention and 33 a 512-token sliding window, a 256K context and three reasoning levels. StepFun publishes official GGUFs from 75.8 GB (IQ3_XXS) to 394 GB (BF16); there is no official Q4_K_M, so the headline figure here is the Q4_K_S file.
Step 3.7 Flash 198B-A11B needs about 112 GB of VRAM at Q4_K_S — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| IQ3_XXS 3.09 bpw | 77.1 GB | — | Too big | |
| Q3_K_M 3.41 bpw | 95.1 GB | — | Too big | |
| Q3_K_L 4.18 bpw | 103.8 GB | — | Too big | |
| IQ4_XS 4.29 bpw | 106.3 GB | — | Too big | |
| Q4_K_S 4.55 bpw | 112.8 GB | — | Too big | |
| Q8_0 8.50 bpw | 210.8 GB | — | Too big | |
| BF16 16.00 bpw | 395.4 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Step 3.7 Flash 198B-A11B VRAM calculator →
Get Step 3.7 Flash 198B-A11B running
The cheapest catalogued GPU that runs Step 3.7 Flash 198B-A11B is the NVIDIA DGX Spark (128 GB).
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run Step 3.7 Flash 198B-A11B
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.
Specifications
Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.
- Parameters
- 198 Billion (11B active)
- Context window
- 262,144
- Architecture
- Mixture-of-Experts (288 routed + 1 shared, 8 active), GQA with 3:1 sliding-window to full attention
- Provider
- StepFun
- Licence
- Apache-2.0
- Practical deployment
- Workstation (large unified memory or pro GPU)
- Specified at
- Q4_K_S
- Vision projector
- +4 GB when you send images (not included above)
- System RAM
- 128 GB
- Record updated
- 2026-11-03
Commercial use permitted. No usage restrictions beyond attribution.
Limits and caveats
- No official Q4_K_M build: the published 4-bit files are IQ4_XS (105.0 GB) and Q4_K_S (111.5 GB).
- StepFun recommends 128 GB of unified memory; at the default Apple Silicon GPU limit only the 3-bit files leave room for a cache.
- Benchmarks on the card are vendor-reported.
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Other StepFun Step sizes
Step 3.7 Flash 198B-A11B — frequently asked questions
How much VRAM does Step 3.7 Flash 198B-A11B need?
About 112 GB at Q4_K_S — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Step 3.7 Flash 198B-A11B run on an RTX 4090 (24 GB)?
No. Step 3.7 Flash 198B-A11B needs about 112 GB at Q4_K_S, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run Step 3.7 Flash 198B-A11B locally?
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. Running the published tag would send your prompts to a hosted GPU rather than your own machine.
What other sizes does StepFun Step come in?
Step 3.7 Flash 198B-A11B (112 GB), Step 5 Preview 600B-A27B (366 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.