StepFun

StepFun (阶跃星辰) develops the Step series of multimodal models with strong performance in image and video understanding. Their models combine language and vision capabilities for comprehensive media analysis and generation tasks.

Official Website Documentation

Endpoints

POST

Chat Completions

/v1/chat/completions

Chat with Step multimodal models.

Streaming

Vision

Function Calling

Models

step-2-16kstep-1v-8kstep-1-8k

Pricing

Model	Input Price	Output Price	Context Window
Step-2 16K	$0.35	$0.35	16.0K
Step-1V	$0.70	$0.70	8.0K

Capabilities

Streaming

Function Calling

Vision

Image Input

Video Input

Long Context

Try it in Playground

Test StepFun APIs directly in our interactive playground with your own API key.

Try it in Playground

StepFun

Endpoints

Chat Completions

Capabilities

Try it in Playground

Overview

StepFun

Endpoints

Chat Completions

Capabilities

Try it in Playground

Overview