Step-Audio 2 Mini Base
StepFun publishes the Base checkpoint separately from the Mini dialogue model. It uses the Step-Audio 2 framework and has a dedicated base-model inference example.
Research release
Use the base example
The repository directs users to examples-base.py. The shared card also describes the finished family, so its dialogue and benchmark claims should not be treated as a separate evaluation of Base.StepFun Step-Audio-2-mini-Base model card
Start with a defined adaptation task
Practical guidance. Use this release when you need to train on a new audio domain or study the base representations. For an assistant that must already follow conversation instructions, the finished Mini checkpoint is a more direct starting point.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | Step-Audio-2-mini-Base |
| License | Apache 2.0 for model and code.StepFun Step-Audio-2-mini-Base model card |
| Stage | Base training checkpoint in the Mini model family.StepFun Step-Audio-2-mini-Base model card |
Related models
- Step-Audio 2 Mini
StepFun · StepAudio 2
A speech-dialogue model supporting sound understanding, tool calls and retrieval from text and acoustic material.
Launch date not confirmedAPI · Downloadable - Step-Audio 2 Mini Think
StepFun · StepAudio 2
A Step-Audio 2 Mini checkpoint for audio understanding and spoken dialogue.
Launch date not confirmedResearch releaseDownloadable