Description
Step3 is a cutting-edge multimodal reasoning model—built on a Mixture-of-Experts architecture with 321B total parameters and 38B active. It is designed end-to-end to minimize decoding costs while delivering top-tier performance in vision–language reasoning. Through the co-design of Multi-Matrix Factorization Attention (MFA) and Attention-FFN Disaggregation (AFD), Step3 maintains exceptional efficiency across both flagship and low-end accelerators.
API Usage Examples
OpenAI Compatible Endpoint
Use this endpoint with any OpenAI-compatible library. Model: StepFun: Step3 (stepfun-ai/step3)
curl https://api.ridvay.com/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer YOUR_API_KEY" -d '{
"model": "stepfun-ai/step3",
"messages": [
{
"role": "user",
"content": "Explain the capabilities of the StepFun: Step3 model"
}
],
"temperature": 0.7,
"max_tokens": 1024
}'Supported Modalities
- Text
- Images
API Pricing
- Input: 0.57$ / 1M tokens
- Output: 1.42$ / 1M tokens
Token Limits
- Max Output: 65,536 tokens
- Max Context: 65,536 tokens
Subscription Tiers
- free
- pro
- ultimate