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...
Specifications
- Provider
- StepFun
- Type
- Open-source / open-weight
- Modality
- text+image+video->text
- Context window
- 262,144
- Released
- May 28, 2026
Capabilities
Input: textInput: imageInput: videoOutput: textReasoning requiredConfigurable reasoning effort
Strengths
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
More from StepFun
