Step 5 Preview

stepfun/step-5-preview

A chat model with a 1M-token context window, at $1.05 per million input tokens and $2.835 per million output tokens.

Details

Maker
stepfun
Kind
Chat
Context window
1,048,576 tokens
Max output
1,048,576 tokens
Input
$1.05 per million tokens
Cached input
$0.105 per million tokens, 90% off
Cache write
Same as input
Output
$2.835 per million tokens
Capabilities
Reasoning, tool use, vision
Takes
Text, image, video
Returns
Text
Zero data retention
Not on every provider
No training on requests
Not on every provider
Released
2026-09-20

Description

Step 5 Preview is StepFun’s flagship AI model for agentic coding, professional knowledge work, and financial analysis. Built on a sparse Mixture-of-Experts architecture with 600B total parameters and 27B active per token, it combines a 1M-token context window with vision input. From building interactive applications and debugging code to conducting research and producing analytical reports, it supports complex workflows that require sustained reasoning, tool use, and iterative refinement.

Call it

curl
curl https://phantom.codes/v1/chat/completions \
  -H "Authorization: Bearer $PHANTOM_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "stepfun/step-5-preview",
    "messages": [
      {"role": "user", "content": "Hello"}
    ]
  }'

Set PHANTOM_KEY to your key. Any OpenAI client works with the base URL https://phantom.codes/v1.