Step 5 Preview: StepFun's frontier model with 1M context and video input

Posted by AnneWodell on Monday September 21, 2026 @09:43AM
from the platform.stepfun.ai dept.

https://platform.stepfun.ai/docs/en/guides/models/step-5-preview

( 1 comment | 2 points | on HN )

Re:Step 5 Preview: StepFun's frontier model with 1M context ... by lowkeysmile on Monday September 21, 2026 @10:16AM (#49787599)

Anyone tried this for actual coding yet? Curious how token-effective it is compared to Kimi K3. The benchmark results looks close but in real-world tasks it may different.

[ Parent | on HN ]