StepFun Step 5 Preview
45 +
Percs
Multi-modal
Support file upload
About
StepFun's flagship model for agentic work: sparse Mixture-of-Experts (27B active / 600B total), strong at software engineering and professional tasks, 1M-token context.
Settings
Response length- The maximum number of tokens to generate in the output.
Temperature- Higher values make the model more creative, lower values more focused.
Diversity control- Top_p. Lower values = top few likely responses, higher values = larger pool.
Context length- The maximum number of tokens to use as input for a model.