Artificial Analysis says StepFun’s StepAudio 3 ASR is available through the StepFun API for non-streaming transcription, scoring 1.7% WER on the AA-WER Index, tied with Alibaba’s Fun-Realtime-ASR-preview and down from 4.7% for StepAudio 2.5 ASR. It transcribes at 88x real time and costs $0.40 per hour.
StepFun says onPanda, its internal tool for LLM data annotation and model inspection, is now open source. It reports a 52% lower median annotation time than manual post-editing and on-policy fidelity of ΔPPL under 1% versus the model’s resampling baseline.
StepFun provides a bash script to install Step Code on macOS, Linux, or WSL. On Windows, it recommends WSL and says PowerShell support is currently in beta.
StepPage publishes a local static site to a shareable URL with one command, with version management and rollback built in so development, debugging, and delivery can stay in the same terminal.
StepFun has released the Step Code v0.1.0 command-line tool as open source under the MIT License, including one-command static site publishing with StepPage. StepFun reports 80.9% on Terminal-Bench 2.1 and 73.3% on Multi-Frame, its 150-task long-horizon benchmark.
Step Code v0.1.0 is an MIT-licensed CLI agent that reads and edits code, runs tests, and ships from one command line. StepFun reports 80.9% on Terminal-Bench 2.1 and 73.3% on Multi-Frame, its 150-task benchmark, and includes one-command publishing with StepPage.
StepFun highlighted Artificial Analysis results putting Step 5 Preview at 44 on the Intelligence Index and $0.71 per task, with more to come on Oct 15. The evaluation describes a 600B/27B MoE flagship with first-party API access and open weights planned for October 15.
elvis (@omarsar0) says he early-tested StepFun’s new Step 5 Preview as a coding agent. He reports that it and GLM 5.3 both solved two real repository tasks with no fix rounds, that Step 5 Preview stopped after checking its work, and that it found five hidden clues in a ~368K-token long-context test.
