Updated daily at 05:00 UTC
Speech-to-text
Qwen-Audio
Open audio-language models that take speech or sound plus text and answer in text.
By Alibaba Group.
Versions
3 versions
| Model | Maker | Kind | Released | Context | Input $/1M | Output $/1M |
|---|---|---|---|---|---|---|
| Qwen3-Omni-30B-A3B-CaptionerQwen-Audio · Open weights | Alibaba Group | Speech-to-text | 22 Sep 2025 | — | $3.81 | $3.06 |
| Qwen2-Audio-7B-InstructQwen-Audio · Legacy · Open weights | Alibaba Group | Speech-to-text | 9 Aug 2024 | 8.2K | — | — |
| Qwen-Audio-ChatQwen-Audio · Legacy · Open weights | Alibaba Group | Speech-to-text | 30 Nov 2023 | 2K | — | — |
