Step-Audio 2 mini
Apache-2.0 fully open speech-to-speech model rivaling commercial voice APIs.
About
StepFun's Step-Audio 2 mini is a fully open-source (Apache 2.0) speech-to-speech model: raw audio in, reasoned audio out, no ASR-LLM-TTS pipeline in the middle. It keeps emotion, accent, and speaking style intact, handles multiple languages, and can call external tools mid-conversation.
StepFun claims it rivals the best commercial voice APIs on accuracy and conversational quality — take that as marketing until you test it. The larger Step-Audio family (R1, EditX) covers reasoning and editing if you need those instead.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
- Rivals best commercial APIs in accuracy and conversational quality (StepFun-reported)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: StepFun release 2025-08-29; Apache-2.0 per third-party coverage; End-to-end speech-to-speech; preserves emotion/accent/style; Commercial-API parity claim is StepFun marketing — untested independently. License: Apache-2.0 per third-party coverage of the release; verify on the repo..
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 6.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 8.0/10
- Manual review · sanity read + red-flag check
- 6.0/10
- · StepFun release 2025-08-29; Apache-2.0 per third-party coverage
- · End-to-end speech-to-speech; preserves emotion/accent/style
- · Commercial-API parity claim is StepFun marketing — untested independently