> ## Documentation Index
> Fetch the complete documentation index at: https://platform.stepfun.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Step 5 Preview

<div className="step-model-doc">
  **Step 5 Preview is StepFun's flagship model for agentic work.** It delivers frontier-level performance across software engineering and professional knowledge work, with particular strength in finance. It natively supports text, image, and video input and provides a **1M-token context window**, making it suitable for tasks that require large amounts of information, tool calls, and continuous progress toward a deliverable.

  ## At a glance

  | Specification      | Step 5 Preview          |
  | :----------------- | :---------------------- |
  | **Model ID**       | `step-5-preview`        |
  | **Context window** | **1M tokens**           |
  | Input types        | Text, images, and video |
  | Output type        | Text                    |
  | Maximum input      | **1M tokens**           |
  | Maximum output     | **1M tokens**           |

  ## Core capabilities

  <Columns cols={2}>
    <Card title="Long-context understanding and reasoning">
      Analyze long documents, multiple source materials, and conversation history within a 1M-token context.

      **Typical tasks**: cross-document question answering and research material organization.
    </Card>

    <Card title="Programming and software engineering">
      Work across multiple programming languages and engineering tasks, with tools to advance development, modification, and verification.

      **Typical tasks**: troubleshooting, code changes, and test creation.
    </Card>

    <Card title="Multi-step Agent tasks">
      Use tools provided by an application to retrieve information, process documents, and continue through multi-step work.

      **Typical tasks**: deep research and analytical reporting.
    </Card>

    <Card title="Multimodal understanding">
      Combine images, video, and text to extract and analyze multimodal information.

      **Typical tasks**: chart analysis, screenshot question answering, and video summarization.
    </Card>
  </Columns>

  ## Use cases

  * **Software development**: Combine code, dependency information, and error logs to locate problems, modify code, and recommend tests.
  * **Long-document processing**: Organize multiple sources in a long context, extract and summarize information, and cross-check details.
  * **Research and analysis**: Work with search, code execution, or document tools to break down tasks and produce structured results.
  * **Multimodal understanding**: Analyze screenshots, charts, and video, and output conclusions in a specified format.

  Tools, search, and code execution capabilities are provided by the application that integrates the model; the model does not itself access your local environment or external services.

  ## Input and output

  ### Image and video input

  | Item               | Supported range                                                       |
  | :----------------- | :-------------------------------------------------------------------- |
  | Image input method | URL or Base64                                                         |
  | Image formats      | JPG/JPEG, PNG, WebP, and static GIF                                   |
  | Images per request | Up to 60 images                                                       |
  | Image detail level | `low`, `high`                                                         |
  | Video input method | URL, Base64, or a `stepfile://` reference from the Files API          |
  | Video formats      | MP4, QuickTime, and Matroska                                          |
  | Video URL limit    | Individual MP4 files under 128 MB; less than 5 minutes is recommended |

  See the [Quickstart](/docs/en/quickstart/overview) for image and video request examples. For complete limits, see [Image understanding best practices](/docs/en/guides/developer/image-chat) and [Video understanding best practices](/docs/en/guides/developer/video-chat).

  ## Features and configuration

  | Feature / configuration | Supported range / parameter                                                                                                                 | Guide                                                                 |
  | :---------------------- | :------------------------------------------------------------------------------------------------------------------------------------------ | :-------------------------------------------------------------------- |
  | Streaming output        | Supported                                                                                                                                   | [Streaming responses](/docs/en/guides/developer/stream)                    |
  | Tool calling            | Supported                                                                                                                                   | [Tool calling](/docs/en/api-reference/tool-call)                           |
  | Structured output       | JSON Mode and JSON Schema supported                                                                                                         | [JSON Mode](/docs/en/guides/developer/json-mode)                           |
  | Reasoning effort        | Supports `low`, `medium`, and `high`; use `reasoning_effort` with the Chat Completions API and `output_config.effort` with the Messages API | [Reasoning models](/docs/en/guides/developer/reasoning)                    |
  | Prompt caching          | Supported                                                                                                                                   | [Prompt caching](/docs/en/guides/developer/prompt-cache)                   |
  | Output length           | `max_tokens` defaults to `INF`; no limit is applied and the model determines it automatically                                               | [Chat Completions API](/docs/en/api-reference/chat/chat-completion-create) |

  ## Integration

  1. Create an API key on the StepFun Open Platform.
  2. Open the [Quickstart](/docs/en/quickstart/overview), select Step 5 Preview, and switch between the text chat, image understanding, and video understanding examples.
  3. Send a request with cURL or Python and read the result from `choices[0].message.content`.
  4. Before going live, use the [API reference](/docs/en/api-reference/chat/chat-completion-create) to add timeout, retry, error-handling, and key-management logic.

  <Columns cols={2}>
    <Card title="Quickstart" href="/docs/en/quickstart/overview">
      See text, image, and video request examples.
    </Card>

    <Card title="Chat Completions API" href="/docs/en/api-reference/chat/chat-completion-create">
      See the complete request parameters, response fields, and usage information.
    </Card>

    <Card title="Claude Code integration" href="/docs/en/step-plan/integrations/claude-code">
      Configure Step Plan and enable 1M context.
    </Card>
  </Columns>

  ## Pricing and rate limits

  Step 5 Preview is billed based on actual input and output token usage. For current prices, cache billing rules, and account rate limits, see [Pricing and Rate Limits](/docs/en/guides/pricing/details#pricing-for-multimodal-reasoning-models).

  ## More information

  [Quickstart](/docs/en/quickstart/overview) · [API reference](/docs/en/api-reference/chat/chat-completion-create) · [Claude Code integration](/docs/en/step-plan/integrations/claude-code) · [Pricing and Rate Limits](/docs/en/guides/pricing/details)
</div>