Step 3.7 Flash
Step 3.7 Flash is StepFun's flagship multimodal reasoning model, a sparse MoE with 11B active parameters and native image and video understanding. It supports a context window of 256K tokens and a max output of 256K tokens per request.
View API reference- Input and output price
- Input $0.20, Output $1.15, Per 1M tokens
- 24h uptime
- Loading AI Gateway uptime
import { streamText } from 'ai'
const result = streamText({ model: 'stepfun/step-3.7-flash', prompt: 'Why is the sky blue?'})Copy link to headingPlayground
Try out Step 3.7 Flash by StepFun. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Step 3.7 Flash
Step 3.7 Flash
Copy link to headingProviders
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
Copy link to headingUptime24 hours
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
Copy link to headingThroughput24 hours
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
Copy link to headingLatency24 hours
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Getting started
Call Step 3.7 Flash through AI Gateway with the AI SDK generateText and streamText functions, or through the OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages APIs by changing the base URL. AI Gateway authenticates the request and routes it to an available provider.
Install the AI SDK (pnpm add ai dotenv), create an API key from the API Keys page, and set it as AI_GATEWAY_API_KEY in your environment. Full setup is covered in the text generation quickstart.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', prompt: 'Why is the sky blue?', });
console.log(result.text);}
main().catch(console.error);Top-level parameters
The same Step 3.7 Flash request in each API format AI Gateway supports.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', system: 'You are a concise technical assistant.', prompt: 'Summarize the tradeoffs between static generation and SSR.', maxOutputTokens: 1024, });
console.log(result.text);}
main().catch(console.error);Standard parameters like prompt, messages, temperature, and tools work as documented in the AI SDK docs. These are the parameters with model-specific behavior.
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Model ID in the form creator/model, e.g. stepfun/step-3.7-flash. AI Gateway routes the request to an available provider. |
maxOutputTokens | number | No | Hard cap on generated tokens. Step 3.7 Flash supports up to 256,000 output tokens. Reasoning tokens count toward this limit. |
reasoning | 'provider-default' | 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | No | Provider-agnostic reasoning effort, available in AI SDK 7 or later. Maps to the provider’s native reasoning configuration; reasoning settings under providerOptions take precedence when both are set. See the Reasoning section below. |
providerOptions | Record<string, JSONValue> | No | AI Gateway routing options under gateway, plus any provider-native options under the provider’s own namespace — see the table below. |
Input limits
| Input | Formats | Sources | Max count | Max size | Limits |
|---|---|---|---|---|---|
| Text | — | — | — | — | Prompt and response share the 256K-token context window |
| Image | — | URL, base64, Uint8Array | — | — | Sent as image parts in messages; counts as input tokens |
Provider options
Set AI Gateway routing options under providerOptions.gateway. For provider-specific options, pass them under the provider’s namespace as documented by the AI SDK.
Learn more in the AI SDK provider docs.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', prompt: 'Why is the sky blue?', providerOptions: { gateway: { only: ['stepfun'], }, }, });
console.log(result.text);}
main().catch(console.error);These AI Gateway routing options apply to every model. Provider-specific options pass through under the provider’s own namespace (for example providerOptions.anthropic) exactly as documented by the AI SDK.
| Parameter | Type | Required | Description |
|---|---|---|---|
providerOptions.gateway.only | string[] | No | Restrict routing to these provider slugs. Requests fail over only within the listed providers. |
providerOptions.gateway.order | string[] | No | Preferred provider order. Listed providers are tried first; unlisted providers remain available as fallbacks. |
providerOptions.gateway.sort | 'cost' | 'ttft' | 'tps' | No | Rank candidate providers by price, time to first token, or tokens per second instead of the default routing order. |
providerOptions.gateway.zeroDataRetention | boolean | No | Route only to providers with a zero-data-retention policy for this model. |
Routing across providers
AI Gateway serves the same model through multiple providers and fails over automatically. order expresses a preference while keeping every provider eligible; only is a hard allowlist — if none of the listed providers are available the request fails instead of falling back.
Options under a provider's own namespace (for example providerOptions.anthropic) are forwarded to that provider with the request. Providers ignore option namespaces that don't apply to them, so it is safe to set provider options alongside gateway routing options.
Reasoning
AI Gateway bridges reasoning across every API format. The AI SDK exposes a provider-agnostic top-level reasoning level (none, minimal, low, medium, high, or xhigh); the Chat Completions and Responses formats take the same effort under reasoning.effort; and the Anthropic Messages format uses a native thinking token budget. Whichever you send, the gateway maps it to the target model’s native configuration, converting between effort levels and token budgets as needed. Reasoning-related settings under providerOptions take full precedence over the top-level reasoning value and are never merged. Reasoning tokens typically count toward your output-token usage, though how they’re reported and billed varies by provider.
Learn more in the AI Gateway reasoning guide.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', prompt: 'Explain the Monty Hall problem step by step.', reasoning: 'high', });
console.log(result.text);}
main().catch(console.error);Image input
Send images alongside text as message parts. Images count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this image.' }, { type: 'image', image: 'https://example.com/photo.jpg' }, ], }, ], });
console.log(result.text);}
main().catch(console.error);Tool calling
Expose tools the model can call. Define each tool’s inputs with a Zod schema.
import { generateText, tool } from 'ai';import { z } from 'zod';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'stepfun/step-3.7-flash', prompt: 'What is the weather in San Francisco?', tools: { getWeather: tool({ description: 'Get the current weather for a location', inputSchema: z.object({ location: z.string() }), execute: async ({ location }) => ({ location, temperatureC: 18 }), }), }, });
console.log(result.text);}
main().catch(console.error);Copy link to headingMore models by StepFun
| Model |
|---|
| 262K | 0.9 s | 155 tps | $0.09/M | $0.30/M | Read$0.02/M | — | +1 |  | 01/29/2026 |
Copy link to headingAbout Step 3.7 Flash
Released on May 28, 2026, Step 3.7 Flash is StepFun's flagship multimodal reasoning model. The architecture pairs the sparse mixture-of-experts language backbone from step-3.5-flash with a dedicated vision encoder, for roughly 198B total parameters and 11B active per token. Weights are published on GitHub, Hugging Face, and ModelScope.
Vision is built for action, not just description. Step 3.7 Flash reads product UIs, documents, charts, spreadsheets, and natural scenes, then writes code or calls tools based on what it sees. A built-in Python tool lets Step 3.7 Flash crop, zoom, and annotate images mid-task, and visual search extends coverage to long-tail entities. That composition shows up in the numbers: 95.29% on V* with Python tooling and 89.13% on HR-Bench 4K.
Agentic performance improves alongside the new modality. Step 3.7 Flash scores 76.5% on SWE-Bench Verified, 56.3% on SWE-Bench Pro, and 59.5% on Terminal-Bench 2.1, gaining about five points on SWE-Bench Pro and six points on Terminal-Bench 2.1 over step-3.5-flash. On search and research tasks, Step 3.7 Flash posts 75.82% on BrowseComp and 47.2% on Humanity's Last Exam with tools. StepFun trained the model to treat search as part of reasoning, favoring search planning and evidence filtering over memorized knowledge.
Through AI Gateway, you call Step 3.7 Flash with one API key and get provider routing, automatic failover, and built-in observability. Integrate via the AI SDK, the Chat Completions API, the Responses API, the Messages API, or other supported API formats.
Copy link to headingWhat To Consider When Choosing a Provider
- Configuration: Step 3.7 Flash spends tokens on both visual analysis and reasoning before answering. Multimodal requests with large images or many reasoning steps consume more output budget than plain text chat, so set token limits with headroom.
- Configuration: Benchmark figures come from StepFun's published evaluations, and some competitor comparisons in the release notes are self-tested. Validate on your own workload before moving production traffic. For current throughput and latency, see live metrics on this page.
- Zero Data Retention: Zero Data Retention is offered on a per-provider and model basis. See the documentation for details.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
Copy link to headingWhen to Use Step 3.7 Flash
Best for
- Multimodal Agent Workflows: Agents that act on screenshots, documents, charts, and UI captures
- Agentic Coding: Software tasks backed by a 76.5% SWE-Bench Verified score and strong terminal control
- Deep Visual Search: Research workflows that combine web browsing with image understanding
- Enterprise Document Work: Finance, data analysis, and mixed-format report processing
- GUI Automation: Phone-use and desktop tasks that navigate real application interfaces
Consider alternatives when
- Text-Only Workloads:
step-3.5-flashhandles pure text agents without the multimodal premium - Maximum Benchmark Headroom: Larger frontier models still lead on the hardest coding and research suites
- Simple Single-Turn Chat: Reasoning and vision tokens add overhead that lightweight conversation doesn't need
Copy link to headingConclusion
Step 3.7 Flash extends the Step flash line from text agents to agents that see. You get native image and video understanding, stronger coding scores, and tool use that spans terminals, browsers, and visual inputs. Route Step 3.7 Flash through AI Gateway for failover, observability, and one key across the whole catalog.
Copy link to headingFrequently Asked Questions
What architecture does Step 3.7 Flash use?
Step 3.7 Flash pairs a sparse mixture-of-experts language backbone with a dedicated vision encoder, totaling roughly 198B parameters with 11B active per token. See https://platform.stepfun.ai/docs/en/guides/models/step-3.7-flash for details.
What visual inputs can Step 3.7 Flash understand?
Step 3.7 Flash natively understands images and video, including product UIs, documents, charts, spreadsheets, and natural scenes. A built-in Python tool lets Step 3.7 Flash crop, zoom, and annotate images during a task.
How does Step 3.7 Flash differ from
step-3.5-flash?Step 3.7 Flash adds a vision encoder for native image and video understanding, which
step-3.5-flashlacks. Step 3.7 Flash also improves agentic coding, gaining about five points on SWE-Bench Pro and six points on Terminal-Bench 2.1 over its predecessor.What is the context window for Step 3.7 Flash?
Step 3.7 Flash supports a context window of 256K tokens and a max output of 256K tokens per request.
How well does Step 3.7 Flash perform on benchmarks?
Step 3.7 Flash scores 76.5% on SWE-Bench Verified, 56.3% on SWE-Bench Pro, 59.5% on Terminal-Bench 2.1, and 75.82% on BrowseComp in StepFun's published evaluations. Vision results include 95.29% on V* with Python tooling and 89.13% on HR-Bench 4K.
How do I call Step 3.7 Flash through AI Gateway?
Use the model identifier
stepfun/step-3.7-flashwith the AI SDK, the Chat Completions API, the Responses API, the Messages API, or another supported API format. You authenticate with an AI Gateway API key, and no StepFun account is needed.Does AI Gateway support Zero Data Retention for Step 3.7 Flash?
Zero Data Retention is not currently available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.