Step 3.7 Flash

step Reasoning Tool Calling Attachments Open Weights Structured Output

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...

Providers 2

Released May 28, 2026

Input Modalities text, image, video

Output Modalities text

Tarsk Use coding

Available Providers (2)

Provider	Model ID	Input Cost	Output Cost	Context	Max Output	Docs
Nvidia	`stepfun-ai/step-3.7-flash`	$0/MTok	$0/MTok	256K	16.4K
OpenRouter	`stepfun/step-3.7-flash`	$0.20/MTok	$1.15/MTok	256K	256K

Capabilities

Reasoning

Tool Calling

Attachments

Open Weights

Structured Output