Step 3.7 Flash VRAM Requirements

Step 3.7 Flash at a glance

Made by

StepFun, China

Released

May 23, 2026

Parameters

201.4B, 11B active

Context window

256K tokens

License

Apache 2.0, commercial use allowed

Reads

Text, images and video

Tool calling

Supported

Reasoning

Adjustable reasoning effort

On Hugging Face

462 likes, 14K downloads last month (opens in a new tab)

On ModelScope

stepfun-ai/Step-3.7-Flash (opens in a new tab)

Runs with

vLLM, SGLang, llama.cpp (Ollama, LM Studio), MLX (Mac) · vLLM recipe (opens in a new tab)

VRAM by precision and context length

Weights plus the KV cache of one request (BF16 cache), in GiB. Add 1 to 3 GiB for the inference engine itself.

Step 3.7 Flash GGUF files

Exact size of each quantization in stepfun-ai/Step-3.7-Flash-GGUF (opens in a new tab), and the smallest GPU that runs it with llama.cpp, Ollama or LM Studio and an 8K-token context.

Read every day from Hugging Face. Split files are added together; vision projectors are left out.

How many GPUs to run Step 3.7 Flash

Fewest GPUs that fit the weights plus one 8K-token request. Data-center GPUs use vLLM (90% of memory usable), the others llama.cpp. Several GPUs split the model with tensor or pipeline parallelism.

8-bit: FP8 with vLLM (INT8 on GPUs without FP8), Q8_0 with llama.cpp. 4-bit: AWQ / GPTQ with vLLM, Q4_K_M with llama.cpp, both counted at the Q4_K_M size.

Cheapest way to run Step 3.7 Flash today

For each GPU: the fewest cards that fit Step 3.7 Flash with one 8K-token request, the most faithful weight format at that count, and the cheapest on-demand price (2026-10-30).

On-demand prices from AWS, Google Cloud, Azure, RunPod, Vast.ai, Oracle Cloud, OVHcloud, Scaleway, Verda, Akash and UpCloud, checked every hour. Serving many users needs more KV cache, so more memory: size it in the calculator. Some links are affiliate links: StudioTV may earn a commission, at no extra cost to you.

AWS, Google Cloud and UpCloud prices: Spare Cores (opens in a new tab), CC BY-SA 4.0 (opens in a new tab), shared under the same license.

Or use Step 3.7 Flash through an API

No GPU to rent: these providers serve Step 3.7 Flash and bill per token. Speeds are measured by Hugging Face for one request.

Prices and speeds from Hugging Face Inference Providers (opens in a new tab), read every day (October 30, 2026). An API wins at low volume; renting GPUs wins when they stay busy: the calculator gives the break-even.

Recommended settings for Step 3.7 Flash

Reasoning. Adjustable reasoning effort. Reasoning tokens take context and KV cache like any other token: leave room for them.

KV cache: how memory grows with context

Architecture: MoE 288 experts + 1 shared · 12 full-attention layers + 33 sliding-window layers (512 tokens), GQA 8 KV × 128 · vision encoder. Its KV cache is 3.6× smaller than if every layer used full attention, because most layers only keep a short window of recent tokens. Only 11B parameters are active per token, but all 201B must sit in memory: VRAM depends on the total, speed on the active part.

Step 3.7 Flash VRAM badge

For a model card or a README: the 4-bit size and the smallest GPU it fits on, linked to this page. Data as JSON: /api/models/step-3-7-flash.json.

[![Step 3.7 Flash VRAM](https://studiotvai.com/badge/step-3-7-flash.svg)](https://studiotvai.com/vram-requirements/step-3-7-flash)

Other models

Same family (sliding-window attention): Gemma 2 9B · Gemma 3 4B · Unlimited-OCR · Gemma 4 12B · Gemma 4 31B · Gemma 4 26B-A4B · gpt-oss 20B · gpt-oss 120B · Llama 4 Scout 17B-16E · Llama 4 Maverick 17B-128E · Kolibri-1 · Recursive Flint · Nemotron Labs Dory 1.0 4B · Step 3.5 Flash · Laguna S 2.1 · Laguna XS 2.1 · Command A+ · Inkling Small · Muse Glimmer 30B · Humanizer · Mellum2.1 12B-A2.5B · Spark-X2.5 4B · Ministral 8B · Gemma 4 E4B · Gemma 4 E2B · Spark-X2.6 4B

Estimates, not guarantees: computed from the official config.json with the same engine as the LLM VRAM Calculator. Real usage depends on your engine version and settings.