Step 3.7 Flash
Step 3.7 Flash is a large language model made by StepFun, added to the catalog on 28 May 2026. It has a context window of 262,144 tokens and costs $0.20 per 1M input tokens and $1.15 per 1M output tokens. It accepts text, images and video and produces text. Typical uses: analyzing image and video inputs, working through reasoning-oriented prompts and producing structured text responses.
Price checked · source: OpenRouter
What Step 3.7 Flash is
StepFun: Step 3.7 Flash is a multimodal model from StepFun. It accepts text, images, and video as inputs and produces text. Its context window is 262,144 tokens, and its maximum output is 230,400 tokens. The catalog describes it as a Mixture-of-Experts model with a language backbone and a vision encoder for native image and video understanding. It also states that the model activates roughly 11 billion parameters. These details identify the model’s stated architecture, supported input types, and output limits, without implying that it generates images or video. The large context window can hold substantial material in a single request, subject to the maximum context and output limits.
Supported capabilities include reasoning, tools, structured outputs, and response formats. This combination suits requests that bring together written prompts and visual material, as well as workflows where a response needs a defined structure or tools are relevant. The model can return text about submitted images or videos, and can produce text in supported structured formats. Its context capacity may be useful when a task includes extended instructions or multiple pieces of source material. Example tasks include describing or analyzing visual inputs, answering questions about submitted visual material, organizing information into structured text, and working through reasoning-oriented prompts. The catalog lists broad support for tools and response formats but does not specify particular tools or schemas, so those details depend on the integration using the model. Video and image are listed as inputs, while text is the listed output.
Key features
- Accepts text, image, and video inputs and produces text.
- Has a 262,144-token context window.
- Maximum output is 230,400 tokens.
- Supports reasoning, tools, structured outputs, and response formats.
- Uses a vision encoder for native image and video understanding.
Typical uses
- Analyzing image and video inputs
- Working through reasoning-oriented prompts
- Producing structured text responses
- Discussing visual material with accompanying text
Step 3.7 Flash pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.20 |
| Output, per 1M tokens | $1.15 |
| Cached input read, per 1M tokens | $0.04 |
No price change since it was added on 28 May 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0004 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0059 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0083 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.032 |
Step 3.7 Flash specifications
| Model id | stepfun/step-3.7-flash |
|---|---|
| Version | step-3.7-flash-20260528 |
| Context window | 262,144 tokensAbout 196,608 English words |
| Max output | 230,400 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 28 May 2026 |
| Input types | text, image, video |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Step 3.7 Flash yet.
Compare Step 3.7 Flash
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.