Can I run Step 3.7 Flash locally?

Yes, the weights are open and hosted on Hugging Face under stepfun-ai/Step-3.7-Flash, but the full model is 198B parameters sparse MoE, so you will need substantial hardware to run it at any useful speed.

What does it cost and how large is the context window?

Input tokens cost $0.20 per million ($0.04 on a cache hit) and output tokens cost $1.15 per million; the context window is 256K tokens.

How does Step 3.7 Flash differ from a standard dense model?

It is a Mixture-of-Experts architecture that activates roughly 11B parameters per token despite a 198B total parameter count, which gives it higher throughput than a comparably capable dense model and lets you dial reasoning depth up or down per request.

Step 3.7 Flash

3679°

#llm #stepfun #vision #video #reasoning #agents

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

StepFun | 262,144 token context | $0.20/M input, $1.15/M output. Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters per token. The model supports a 256K context window and exposes selectable reasoning levels (high/medium/low), letting callers trade off speed, cost, and depth of reasoning. Designed for coding, agentic workflows, structured outputs, and long-context productivity tasks.

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

Open weights vs open source vs API-only: what you are actually allowed to do

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Step 3.7 Flash is StepFun's sparse Mixture-of-Experts vision-language model, pairing a 196B-parameter language backbone with a vision encoder while activating only around 11B parameters per token. It supports a 256K context window, native image and video understanding, and three selectable reasoning levels. At $0.20 per million input tokens it targets developers running high-throughput agentic pipelines, not casual chat.

Known for

Selectable reasoning depth (low/medium/high) to trade speed against cost per call.
Strong agentic tool use: leads ClawEval-1.1 at 67.1, well clear of the next competitor.
Native image and video understanding built into the base model, not bolted on.
256K context handles large documents, long codebases, or multi-turn agent histories in one pass.
Open weights available on Hugging Face for self-hosted or on-prem deployment.

Watch out for

At 198B total parameters, running this locally demands serious hardware; check your VRAM before downloading.
Output pricing at $1.15/M tokens adds up fast in high-reasoning, verbose workflows.
StepFun has not published full licence terms in the model card; verify permitted uses before production deployment.

Best for

Prompts to try

Multi-file bug hunt

Here is a GitHub issue report and three related Python files. Trace the root cause across the files, explain exactly which line introduces the bug, and produce a minimal patch that would pass a regression test. Issue: calling process_batch() with an empty list raises a KeyError on 'results'. Files: [paste your files here].

Chart to structured data

I am attaching a screenshot of a financial dashboard with multiple charts and a summary table. Extract every numeric value you can read, identify the metric name and time period for each, and return the result as a JSON array with keys: metric, period, value, unit.

Long-document agent plan

You have a 200-page procurement contract pasted below. Identify every clause that creates a financial obligation on our side, the deadline or trigger for each, and the maximum liability amount if stated. Return a table sorted by deadline date. If a deadline is ambiguous, flag it. [paste contract text].

Compare with

Step 5 Preview
Open-weights model
1,048,576 token context · $1.00/M input, $2.70/M output
Step 3.5 Flash
Open-weights model
262,144 token context · $0.10/M input, $0.30/M output
GPT-6 Luna
API-only model
1,050,000 token context · $0.10/M input, $0.50/M output
GPT-5.6 Luna
API-only model
1,050,000 token context · $0.20/M input, $1.20/M output

More text models from stepfun

Questions people ask

Can I run Step 3.7 Flash locally?: Yes, the weights are open and hosted on Hugging Face under stepfun-ai/Step-3.7-Flash, but the full model is 198B parameters sparse MoE, so you will need substantial hardware to run it at any useful speed.
What does it cost and how large is the context window?: Input tokens cost $0.20 per million ($0.04 on a cache hit) and output tokens cost $1.15 per million; the context window is 256K tokens.
How does Step 3.7 Flash differ from a standard dense model?: It is a Mixture-of-Experts architecture that activates roughly 11B parameters per token despite a 198B total parameter count, which gives it higher throughput than a comparably capable dense model and lets you dial reasoning depth up or down per request.

Captain's verdict
A serious agentic workhorse with genuine vision chops and a clever reasoning dial, provided you can stomach the output cost or host it yourself.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

NVIDIA Nemotron 3 Ultra 550B A55B

TEXT / LLM

AI Models for Image, Video & Text Generation

Step 3.7 Flash is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm stepfun vision video reasoning agents coding long context

Browse by Model Family