Featherless serves StepFun's open-weights Step 5 Preview at FP8
Overview
Featherless now serves StepFun's open-weights Step 5 Preview, a 600B sparse mixture-of-experts model with 27B parameters active per token, at FP8 precision, according to StepFun's announcement on X. StepFun says the model matches Kimi K3 on the Artificial Analysis Intelligence Index and costs about one-sixth as much per task as Claude Opus 5.5, citing Artificial Analysis. Featherless describes the model as built for long-horizon agent work, including software engineering and finance.
Written by AI from the articles below · updated Oct 15, 11:45 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
StepFun@StepFun_aiOfficialStepFun's Step 5 Preview is live on FeatherlessAIStepFun's open-weights Step 5 Preview, a 600B sparse MoE model with 27B active per token, is now served at FP8 on Featherless. StepFun says it matches Kimi K3 on the Artificial Analysis Intelligence Index and costs about one-sixth as much per task as Claude Opus 5.5, per Artificial Analysis. Featherless describes the model as built for long-horizon agent work, including software engineering and finance.
Heat trend
Not enough continuous observations to show a trend yet.