Messages API

Call Step models using an Anthropic Messages API-compatible format, with support for streaming, tool calls, image input, and reasoning effort configuration.

Endpoint

Claude Code and the Anthropic SDK use the configuration Base URL and append /v1/messages. Use the full request URL with curl. For tool setup, see the Claude Code integration guide.

Channel	Configuration Base URL	Full request URL
Standard API	`https://api.stepfun.ai`	`POST https://api.stepfun.ai/v1/messages`
Step Plan	`https://api.stepfun.ai/step_plan`	`POST https://api.stepfun.ai/step_plan/v1/messages`

Request Parameters

The request supports the following top-level fields:

Field	Type	Required	Description
`model`	string	Yes	Model ID, such as `step-5-preview`, `step-3.7-flash`, or `step-3.5-flash`. `step-router-v1` is available only through Step Plan; see its field restrictions below.
`messages`	object array	Yes	Conversation messages; at least one. Each message contains `role` (`user` / `assistant`) and `content` (plain text or a content block array).
`max_tokens`	int	Yes	Maximum number of generated tokens; must be greater than 0.
`system`	string or array	No	System prompt, as a string or an array of text blocks.
`tools`	object array	No	Tool definitions. Each entry contains `name`, `description`, and `input_schema` (JSON Schema).
`output_config`	object	No	Output configuration. `effort` controls reasoning depth: models supporting three levels accept `low` / `medium` / `high`; `step-3.5-flash-2603` accepts `low` and `high`.
`stream`	boolean	No	Whether to stream the response. Defaults to non-streaming.
`temperature`	float	No	Sampling temperature, from 0 to 2.
`top_p`	float	No	Nucleus sampling parameter, greater than 0 and at most 1.
`top_k`	int	No	Top-k parameter, from 0 to 500.
`stop_sequences`	string array	No	Generation stops when any sequence in this list appears in the output.

Message Content Format

The content field of each entry in messages accepts plain text (a string) or an array of the following content blocks:

Block	type	Fields
Text	`text`	`text`: text content.
Image	`image`	`source` supports a URL, `{"type":"url","url":"https://..."}`, or Base64, `{"type":"base64","media_type":"image/png","data":"..."}`.
Tool call (from the model)	`tool_use`	`id`: unique identifier; `name`: tool name; `input`: argument object.
Tool result (from the caller)	`tool_result`	`tool_use_id`: matching call ID; `content`: execution result; `is_error`: whether execution failed.

step-router-v1 Field Restrictions

step-router-v1 is available only through Step Plan and routes between deepseek-v4-pro and step-3.7-flash. The restrictions below apply only to this model; other fields follow the request parameters above.

Field	Behavior when calling step-router-v1
`model`	The router model ID is `step-router-v1`; an invalid router model name returns HTTP 400 `request_params_invalid`.
`max_tokens`	Maximum: 250K.
Image / document blocks in `messages.content`	Not supported; returns `unsupported_content_type`.
`web_search` in `tools`	Not supported; returns `unsupported_content_type`.
`output_config.effort`	Ignored.

Response

Non-streaming response

Content-Type: application/json

{
    "id": "msg_xxx",
    "type": "message",
    "role": "assistant",
    "model": "step-5-preview",
    "stop_reason": "end_turn",
    "usage": {
        "input_tokens": 20,
        "output_tokens": 12
    },
    "content": [
        {
            "type": "text",
            "text": "I'm an AI assistant."
        }
    ]
}

Response Fields

Field	Type	Description
`id`	string	Unique message identifier.
`type`	string	Object type, always `message`.
`role`	string	Role, always `assistant`.
`model`	string	Model ID used for this response.
`content`	object array	Response blocks, which may include `text`, `thinking`, or `tool_use`.
`stop_reason`	string	Why generation stopped: `end_turn`, `tool_use`, or `max_tokens`.
`usage`	object	Token usage, including `input_tokens` and `output_tokens`. Cache usage fields may also be present.

Streaming response

Content-Type: text/event-stream Streaming uses standard SSE format. Each event has an event: line and a data: line; data: is JSON. Common event types: message_start, content_block_start, content_block_delta, content_block_stop, message_delta, message_stop, ping. When streaming tool-call arguments, content_block_delta.delta.type may be input_json_delta.

event: message_start
data: {"type":"message_start","message":{"id":"msg_xxx","type":"message","role":"assistant","model":"step-5-preview","content":[],"stop_reason":null,"stop_sequence":null,"usage":{"input_tokens":20,"output_tokens":1}}}

event: content_block_start
data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"Hello"}}

event: content_block_stop
data: {"type":"content_block_stop","index":0}

event: message_delta
data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{"output_tokens":12}}

event: message_stop
data: {"type":"message_stop"}

Examples

Replace YOUR_STEP_API_KEY with your key. Set STEP_API_KEY before running examples that use the environment variable, then select the standard API or Step Plan tab for your channel.

Basic chat
Streaming response
Using output_config.effort

Standard API
Step Plan

from anthropic import Anthropic

client = Anthropic(api_key="YOUR_STEP_API_KEY", base_url="https://api.stepfun.ai")

message = client.messages.create(
    model="step-5-preview",
    max_tokens=1024,
    system="You are an AI chat assistant provided by StepFun. You are fluent in English, Chinese, and many other languages. You answer user questions quickly and accurately while protecting user data.",
    messages=[
        {
            "role": "user",
            "content": "Introduce yourself in one sentence."
        }
    ],
)

print(message)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
    apiKey: "YOUR_STEP_API_KEY",
    baseURL: "https://api.stepfun.ai"
});

async function main() {
    const message = await client.messages.create({
        model: "step-5-preview",
        max_tokens: 1024,
        system: "You are an AI chat assistant provided by StepFun. You are fluent in English, Chinese, and many other languages. You answer user questions quickly and accurately while protecting user data.",
        messages: [
            {
                role: "user",
                content: "Introduce yourself in one sentence."
            }
        ]
    });

    console.log(JSON.stringify(message));
}

main();

curl https://api.stepfun.ai/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $STEP_API_KEY" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "system": "You are an AI chat assistant provided by StepFun. You are fluent in English, Chinese, and many other languages. You answer user questions quickly and accurately while protecting user data.",
    "messages": [
        {
            "role": "user",
            "content": "Introduce yourself in one sentence."
        }
    ]
}'

{
    "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
    "type": "message",
    "role": "assistant",
    "model": "step-5-preview",
    "stop_reason": "end_turn",
    "usage": {
        "input_tokens": 35,
        "output_tokens": 20
    },
    "content": [
        {
            "type": "text",
            "text": "I'm an AI chat assistant by StepFun, ready to answer your questions in English, Chinese, and other languages quickly and accurately."
        }
    ]
}

import os
from anthropic import Anthropic

client = Anthropic(
    api_key=os.environ["STEP_API_KEY"],
    base_url="https://api.stepfun.ai/step_plan",
)

message = client.messages.create(
    model="step-5-preview",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Reply with OK."
        }
    ],
)

print(message.content)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
    apiKey: process.env.STEP_API_KEY,
    baseURL: "https://api.stepfun.ai/step_plan"
});

async function main() {
    const message = await client.messages.create({
        model: "step-5-preview",
        max_tokens: 1024,
        messages: [
            {
                role: "user",
                content: "Reply with OK."
            }
        ]
    });

    console.log(message.content);
}

main();

export STEP_API_KEY="YOUR_STEP_API_KEY"

curl https://api.stepfun.ai/step_plan/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${STEP_API_KEY}" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "messages": [
        {
            "role": "user",
            "content": "Reply with OK."
        }
    ]
}'

{
    "id": "msg_xxx",
    "type": "message",
    "role": "assistant",
    "model": "step-5-preview",
    "stop_reason": "end_turn",
    "usage": {
        "input_tokens": 20,
        "output_tokens": 1
    },
    "content": [
        {
            "type": "text",
            "text": "OK"
        }
    ]
}

Standard API
Step Plan

from anthropic import Anthropic

client = Anthropic(api_key="YOUR_STEP_API_KEY", base_url="https://api.stepfun.ai")

with client.messages.stream(
    model="step-5-preview",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Introduce yourself in one sentence."
        }
    ],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

print()

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
    apiKey: "YOUR_STEP_API_KEY",
    baseURL: "https://api.stepfun.ai"
});

async function main() {
    const stream = client.messages.stream({
        model: "step-5-preview",
        max_tokens: 1024,
        messages: [
            {
                role: "user",
                content: "Introduce yourself in one sentence."
            }
        ]
    });

    for await (const event of stream) {
        if (
            event.type === "content_block_delta" &&
            event.delta.type === "text_delta"
        ) {
            process.stdout.write(event.delta.text);
        }
    }

    console.log();
}

main();

curl https://api.stepfun.ai/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $STEP_API_KEY" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "stream": true,
    "messages": [
        {
            "role": "user",
            "content": "Introduce yourself in one sentence."
        }
    ]
}'

import os
from anthropic import Anthropic

client = Anthropic(
    api_key=os.environ["STEP_API_KEY"],
    base_url="https://api.stepfun.ai/step_plan",
)

with client.messages.stream(
    model="step-5-preview",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Reply with OK."
        }
    ],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

print()

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
    apiKey: process.env.STEP_API_KEY,
    baseURL: "https://api.stepfun.ai/step_plan"
});

async function main() {
    const stream = client.messages.stream({
        model: "step-5-preview",
        max_tokens: 1024,
        messages: [
            {
                role: "user",
                content: "Reply with OK."
            }
        ]
    });

    for await (const event of stream) {
        if (
            event.type === "content_block_delta" &&
            event.delta.type === "text_delta"
        ) {
            process.stdout.write(event.delta.text);
        }
    }

    console.log();
}

main();

export STEP_API_KEY="YOUR_STEP_API_KEY"

curl https://api.stepfun.ai/step_plan/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${STEP_API_KEY}" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "stream": true,
    "messages": [
        {
            "role": "user",
            "content": "Reply with OK."
        }
    ]
}'

Standard API
Step Plan

from anthropic import Anthropic

client = Anthropic(api_key="YOUR_STEP_API_KEY", base_url="https://api.stepfun.ai")

message = client.messages.create(
    model="step-5-preview",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Explain reinforcement learning in three sentences."
        }
    ],
    extra_body={
        "output_config": {
            "effort": "medium"
        }
    }
)

print(message)

curl https://api.stepfun.ai/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $STEP_API_KEY" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "messages": [
        {
            "role": "user",
            "content": "Explain reinforcement learning in three sentences."
        }
    ],
    "output_config": {
        "effort": "medium"
    }
}'

import os
from anthropic import Anthropic

client = Anthropic(
    api_key=os.environ["STEP_API_KEY"],
    base_url="https://api.stepfun.ai/step_plan",
)

message = client.messages.create(
    model="step-5-preview",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": "Explain reinforcement learning in three sentences."
        }
    ],
    extra_body={
        "output_config": {
            "effort": "medium"
        }
    }
)

print(message)

export STEP_API_KEY="YOUR_STEP_API_KEY"

curl https://api.stepfun.ai/step_plan/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${STEP_API_KEY}" \
  -d '{
    "model": "step-5-preview",
    "max_tokens": 1024,
    "messages": [
        {
            "role": "user",
            "content": "Explain reinforcement learning in three sentences."
        }
    ],
    "output_config": {
        "effort": "medium"
    }
}'

Chat

Realtime

Audio

Image

Models

Files

Account

Tool Call

Token Count

Error Codes

Endpoint

Request Parameters

Message Content Format

step-router-v1 Field Restrictions

Response

Non-streaming response

Response Fields

Streaming response

Examples

​Endpoint

​Request Parameters

​Message Content Format

​step-router-v1 Field Restrictions

​Response

​Non-streaming response

​Response Fields

​Streaming response

​Examples

Endpoint

Request Parameters

Message Content Format

step-router-v1 Field Restrictions

Response

Non-streaming response

Response Fields

Streaming response

Examples