| 1 | # Qwen2.5-Omni Offline Inference Examples |
| 2 | |
| 3 | This folder provides several example scripts on how to inference Qwen2.5-Omni offline. |
| 4 | |
| 5 | ## Thinker Only |
| 6 | |
| 7 | ```bash |
| 8 | # Audio + image + video |
| 9 | python examples/generate/multimodal/qwen2_5_omni/only_thinker.py \ |
| 10 | -q mixed_modalities |
| 11 | |
| 12 | # Read vision and audio inputs from a single video file |
| 13 | python examples/generate/multimodal/qwen2_5_omni/only_thinker.py \ |
| 14 | -q use_audio_in_video |
| 15 | |
| 16 | # Multiple audios |
| 17 | python examples/generate/multimodal/qwen2_5_omni/only_thinker.py \ |
| 18 | -q multi_audios |
| 19 | ``` |
| 20 | |
| 21 | This script will run the thinker part of Qwen2.5-Omni, and generate text response. |
| 22 | |
| 23 | You can also test Qwen2.5-Omni on a single modality: |
| 24 | |
| 25 | ```bash |
| 26 | # Process audio inputs |
| 27 | python examples/generate/multimodal/audio_language_offline.py \ |
| 28 | --model-type qwen2_5_omni |
| 29 | |
| 30 | # Process image inputs |
| 31 | python examples/generate/multimodal/vision_language_offline.py \ |
| 32 | --modality image \ |
| 33 | --model-type qwen2_5_omni |
| 34 | |
| 35 | # Process video inputs |
| 36 | python examples/generate/multimodal/vision_language_offline.py \ |
| 37 | --modality video \ |
| 38 | --model-type qwen2_5_omni |
| 39 | ``` |