Model Catalog › stepfun

StepAudio 2.5 ASR

stepaudio-2.5-asr

TRANSCRIPTION Transcription Proprietary (API-only)

StepAudio 2.5 ASR (stepaudio-2.5-asr) is a transcription model from stepfun; proprietary (API-only); input audio; output text.

Specifications

Weights
Proprietary (API-only)
Input modalities
audio
Output modalities
text
Release date
2026-04-24

Provenance

models-dev
Last verified: 2026-10-30

Related models

Qwen3-ASR Flash
qwen3-asr-flash
qwen · TRANSCRIPTION · 53.2K ctx
gpt-4o-mini-transcribe
gpt-4o-mini-transcribe
openai · TRANSCRIPTION · 16K ctx
gpt-4o-mini-transcribe-2025-03-20
gpt-4o-mini-transcribe-2025-03-20
openai · TRANSCRIPTION · 16K ctx
gpt-4o-mini-transcribe-2025-12-15
gpt-4o-mini-transcribe-2025-12-15
openai · TRANSCRIPTION · 16K ctx
gpt-4o-transcribe
gpt-4o-transcribe
openai · TRANSCRIPTION · 16K ctx
gpt-4o-transcribe-diarize
gpt-4o-transcribe-diarize
openai · TRANSCRIPTION · 16K ctx

Markdown · Vendor JSON · Full catalog · All stepfun models