FFree
StepfunOpen Weights

Step 3.7 Flash

stepfun-ai/Step-3.7-Flash

Best Input Price

$0.00 / 1M

Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters per token. The model supports a 256K context window and exposes selectable reasoning levels (high/medium/low), letting callers trade off speed, cost, and depth of reasoning. Designed for coding, agentic workflows, structured outputs, and long-context productivity tasks.

Context Window

262,144 tokens

Reasoning

Supported

Tool Calling

Supported

Released

2026-05-29

Inference Providers (20)

ProviderModel IDContextInput / 1MOutput / 1MAction
StepFun Step Plan (Global)step-3.7-flash256,000$0.00$0.00Docs ↗
Kilo Gatewaystepfun/step-3.7-flash:free262,144$0.00$0.00Docs ↗
StepFun Step Plan (China)step-3.7-flash256,000$0.00$0.00Docs ↗
UnoRouterstep-3.7-flash:free256,000$0.00$0.00Docs ↗
Kenaristep-3-7-flash:free256,000$0.00$0.00Docs ↗
StepFun (Global)step-3.7-flash256,000$0.18$1.11Docs ↗
StepFun (China)step-3.7-flash256,000$0.18$1.11Docs ↗
Ambientstepfun/step-3.7-flash262,144$0.19$1.14Docs ↗
Deep Infrastepfun-ai/Step-3.7-Flash262,144$0.20$1.15Docs ↗
EmpirioLabs AIstep-3-7-flash256,000$0.20$1.15Docs ↗
NanoGPTstepfun/step-3.7-flash:thinking262,144$0.20$1.15Docs ↗
LLM Gatewaynovita/step-3.7-flash262,144$0.20$1.15Docs ↗
OpenRouterstepfun/step-3.7-flash262,144$0.20$1.15Docs ↗
ZenMuxstepfun/step-3.7-flash256,000$0.20$1.15Docs ↗
Hugging Facestepfun-ai/Step-3.7-Flash262,144$0.20$1.15Docs ↗
Vercel AI Gatewaystepfun/step-3.7-flash256,000$0.20$1.15Docs ↗
Eden AIdeepinfra/stepfun-ai/Step-3.7-Flash262,144$0.20$1.15Docs ↗
Requestystep-3.7-flash262,144$0.20$1.15Docs ↗
DevPass (LLM Gateway)step-3.7-flash262,144$0.20$1.15Docs ↗
AIHubMixstep-3.7-flash256,000$0.22$1.32Docs ↗