HTTP 200 verified twice
Finding your next stop…
Loading the latest directory information.
Loading the latest directory information.
Step 3.7 Flash is a production-grade Agent Flash model on the StepFun platform with enhanced web/visual search and reliable tool orchestration for long-horizon multi-turn Agent workflows. The platform also provides Step Image Edit 2, a unified text-to-image and image editing model.
Start here for the decision-making essentials: what Step-2 does, who it is for, how it is accessed, and the first-party sources behind this profile.
Multimodal images billed at 400 tokens each in addition to text token usage (input + outpu
Read-only public captures of Step-2’s homepage. Screenshots are dated, never live embeds, and open full-screen.
Published specifications and use cases for Step-2. Repository dates are kept separate from the model's original release date.
Short answers to the questions buyers and builders commonly ask about Step-2. Each answer cites the shared ledger below, where every source is listed once.
Web and visual search enhancement that gathers and cross-checks multi-source evidence acro
强化联网检索与图像搜索,使模型在开放信息环境中跨文本与图像主动获取并交叉比对多源证据
API rate limits applied to prevent abuse, ensure fair access, and manage infrastructure lo
速率限制是 API 的常见做法
Multimodal images billed at 400 tokens each in addition to text token usage (input + outpu
按照图片的数量进行按每个图片400Token 消耗进行核算，最终按 Token 的总用量进行计费
Image generation endpoints: POST /v1/images/generations (text-to-image from prompt) and PO
POST /v1/images/generations根据 Prompt 从零生成图像
Compatible with mainstream Agent frameworks including Claude Code, KiloCode, Hermes Agent,
针对主流 Agent 框架（Claude Code / KiloCode / Hermes Agent / OpenClaw等）
This profile connects the jobs Step-2 is described as handling with its delivery model, access options and the subjects used to match it to related products in this directory.
Each fact points to a recorded source, making it easy to distinguish verified product information from claims that need checking.
HTTP 200 verified twice
Step 3.7 Flash 面向生产级 Agent 的高效率 Flash 模型
StepFun platform offers the Step API for models and AI applications · Step 3.7 Flash is a production-grade Agent Flash model · Model is targeted at production-grade Agent applications
StepFun platform offering the Step API for models and AI applications
Step 3.7 Flash — production-grade Agent Flash model
Production-grade Agent applications
Web and visual search enhancement that gathers and cross-checks multi-source evidence across text and images
Reliable tool calling and orchestration across APIs, browsers, terminals, Office tools, and external systems in long-horizon multi-turn Agent workflows
Compatible with mainstream Agent frameworks including Claude Code, KiloCode, Hermes Agent, and OpenClaw
MCP and Skills tool-calling protocol support
Multimodal images billed at 400 tokens each in addition to text token usage (input + output)
API rate limits applied to prevent abuse, ensure fair access, and manage infrastructure load
A concise view of the jobs, capabilities and integrations described in the recorded product sources.
Documented product formats, platforms and official distribution destinations. Availability can vary by region and plan.
A concise history of software releases and material product changes. Events appear only when a dated source supports what changed.
This profile is being checked for dated releases and material product changes. Nothing appears here until the exact date and event can be verified from a recorded source.
Facts, answers, structured details, milestones and primary resource links cite this shared ledger. Each external page appears once; release tags from the same GitHub project are grouped under one release history.
These tools share a workflow, capability or audience with Step-2. They may complement it rather than replace it, and are not presented as integrations or endorsements.
Step Image Edit 2 — lightweight unified text-to-image and image editing model with 1–2 second response
Image generation endpoints: POST /v1/images/generations (text-to-image from prompt) and POST /v1/images/edits (image editing from input image and prompt)
StepFun (阶跃星辰) operates the open platform · Step Plan uses unified Credit unit: 1M Credit = ¥1 · Monthly Credit pool issued once, expires end of month (no rollover)
Native multimodal understanding and execution of UI, charts, documents, images, and application interfaces, converting complex visual information into structured results, code generation, and executable tasks.
Web search and visual search enhancement enabling models to actively retrieve and cross-reference multi-source evidence across text and image in open information environments.
Reliable tool calling and orchestration in long multi-turn Agent workflows, stably invoking APIs, browsers, terminals, Office tools, and external systems while maintaining task trajectory consistency.
Compatibility optimization with mainstream Agent frameworks (Claude Code / KiloCode / Hermes Agent / OpenClaw) and tool-calling protocols including MCP/Skills.
Step Plan uses a unified Credit unit where 1M Credit = ¥1 (i.e., 1,000,000 Credits correspond to ¥1 of model usage).
Step Plan credits are issued once per month into a monthly pool, can be consumed at any time during the month, and do not roll over to the next cycle (expire end of month).
Multimodal API calls are billed at 400 Tokens per image in addition to text token usage.
Standard API fees are calculated as Token usage × model unit price, deducted in order from gift accounts then recharge accounts.
Step Plan requires a dedicated Base URL of https://api.stepfun.com/step_plan/v1 (distinct from the standard API Base URL https://api.stepfun.com/v1); using the wrong URL causes request errors.
Step Plan can be used in agent tools (OpenClaw, Claude Code) and coding tools (Cline, Kilo Code, Roo Code) and other mainstream tools.
Rate limits are applied to the API to prevent abuse/misuse, ensure fair access across users, and manage overall infrastructure load.
Step Plan currently supports the following flagship models: step-3.7-flash (flagship multimodal reasoning), step-3.5-flash (high-speed agent and code), step-3.5-flash-2603 (high-frequency Agent optimization), stepaudio-2.5-realtime (end-to-
StepFun 开放平台 (StepFun Open Platform)
OpenAI SDK compatible Python SDK reuse
step-2x-large | 0.1 元/张 | 文生图 Step 1X image generation, free trial ended 2026-06-12
Step 3.7 Flash — 阶跃星辰旗舰多模态推理模型
198B total / 11B active parameters (Step 3.7 Flash)
Native multimodal input — text, image, and video understanding
Three reasoning effort levels: low / medium / high (Step 3.7 Flash)
Tool calling / function calling supported
JSON structured output supported
step-3.7-flash
Sparse MoE with 198B total / 11B active parameters
256K tokens
Reliable tool calling with multi-step task decomposition and plan execution
