Step-3.7-Flash is an open-weight multimodal model developed by Stepfun AI. It supports vision and text inputs and is distributed on Hugging Face for use with Transformers and local inference engines. The model includes a specialized chat template for handling system prompts, tools, and multimodal content.
Step 3.7 Flash sits in PulseGate's Multimodal & vision category. It focuses on running a fast, multimodal AI model locally or via inference providers without relying on proprietary APIs. Step 3.7 Flash is an open-source project aimed at developers. Step 3.7 Flash is open source under the Open Source license. Step 3.7 Flash is available on the web and the command line.
It is developed by stepfun-ai, and it first shipped in 2025. Key capabilities include Multimodal Input, Chat Template, and Vision Support.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do