Step-Audio 2 is an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation.
active 2025-07-25 → 2026-05-25 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 307 days from 2025-07-23 to 2026-05-25. Pushes: 19 total, peak 7 in a day. Pull requests: 8 total, peak 2 in a day. Issues: 62 total, peak 7 in a day. Comments: 98 total, peak 11 in a day. Stars: 754 total, peak 115 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| petronny | 57 | 13 | 1 | 34 |
| lilyzlt | 7 | 0 | 0 | 1 |
| yangxueruivs | 5 | 1 | 0 | 4 |
| SimZhou | 5 | 0 | 0 | 5 |
| weedge | 4 | 0 | 2 | 1 |
| runninging | 4 | 0 | 0 | 4 |
| dadafsfsdf | 4 | 0 | 0 | 1 |
| wulaoshi | 3 | 0 | 0 | 1 |
| JinmingChe | 3 | 0 | 0 | 1 |
| heheda166 | 3 | 0 | 0 | 2 |
| e1ijah1 | 3 | 0 | 0 | 3 |
| FlyTia | 3 | 0 | 0 | 3 |
| vlordier | 3 | 0 | 2 | 1 |
| Qoboty | 3 | 0 | 0 | 2 |
| ranck626 | 3 | 0 | 0 | 2 |
| Sidak08 | 2 | 0 | 0 | 2 |
| yuekaizhang | 2 | 0 | 1 | 1 |
| PressEtoRace | 2 | 0 | 0 | 1 |
| GallonDeng | 2 | 0 | 0 | 1 |
| levi-li-PG | 2 | 0 | 0 | 2 |
Recent activity
Latest issues, pull requests and releases
- Issue#86npathak132026-03-01 07:07Step-Audio-2-mini running on AMD RDNA 4 (gfx1151 / Strix Halo) via ROCm — build fixes and Dockerfile
- Issue comment#84zh199909062026-02-22 02:22Add asynchronous support for audio processing
- Issue#83nano-micro2026-01-22 03:07推理输出的时候只有文本没有语音
- Issue#82LiXinYuECNU2026-01-02 06:03关于sft step-audio2系列模型的训练数据格式
- Issue#81Pangkaiyuyu2025-12-24 10:24ASR LORA
- Issue#80shanhaidexiamo2025-12-16 09:46Floating point exception (core dumped)
- Issue#79jujunchen2025-12-05 07:05gradio 的版本是多少？
- Issue#55ZhikangNiu2025-12-02 02:36Is the think process of the think model limited to Chinese only?
- Issue comment#70yuekaizhang2025-11-21 06:57Add TensorRT Token2wav
- Issue#76mingyi4562025-11-16 12:44Flash attention not supported?
- Issue#75chenggangqcg2025-11-07 08:44模型输出：无限个嗯，嗯，嗯，嗯，嗯，嗯，嗯
- Issue#71hashiting2025-11-03 08:56流式语音输出
- Issue comment#71petronny2025-10-31 11:56流式语音输出
- Issue comment#58SimZhou2025-10-22 10:44Performance issues
- Issue comment#58SimZhou2025-10-22 10:32Performance issues
- Issue comment#58SimZhou2025-10-22 10:04Performance issues
- Issue comment#58petronny2025-10-22 09:58今天一天测试（对接了FreeSwitch上），试了 三种模型 step-1o-audio ,step-audio-2,step-audio-2-mini 都不行。
- Issue comment#58SimZhou2025-10-22 09:51今天一天测试（对接了FreeSwitch上），试了 三种模型 step-1o-audio ,step-audio-2,step-audio-2-mini 都不行。
- Issue comment#58petronny2025-10-22 09:42今天一天测试（对接了FreeSwitch上），试了 三种模型 step-1o-audio ,step-audio-2,step-audio-2-mini 都不行。
- Issue comment#58SimZhou2025-10-22 09:06今天一天测试（对接了FreeSwitch上），试了 三种模型 step-1o-audio ,step-audio-2,step-audio-2-mini 都不行。
- Pull request#70yuekaizhang2025-10-22 03:02
- Issue#69lilyzlt2025-10-15 09:06How to Support Longer Audio in the vLLM Implementation?
- Issue#67Hooyoung-for-AI2025-10-08 14:59微调代码请求 finetune code request
- Issue#66wwfcnu2025-09-30 03:22vllm环境
- Issue comment#63petronny2025-09-29 03:24vllm版本只支持API推理吗？
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 754 stars here means stars gained during the window, not the repo's star count.