#60 · Primary category： Foundation Models
Qwen2.5-Omni
Qwen2.5-Omni is an end-to-end multimodal model by Qwen team at Alibaba Cloud, capable of understanding text, audio, vision, video, and performing real-time speech generation.
Project last updated：06/12/25
GitHub Stars
4.1K
Forks
327
Contributors
12
License
Apache-2.0
Why we included this project
Qwen2.5-Omni's Thinker-Talker architecture ties the generated speech directly to what the model is currently seeing and hearing, which is why live conversation feels responsive rather than stilted. The model accepts text, images, audio, and video as input and streams back both written answers and synthesized speech, so you can prototype a voice assistant without hooking up separate transcription, vision, and TTS services. The repo includes transformers and vLLM usage examples, 4-bit quantized versions, and notes for edge deployment via MNN, giving a realistic path from a local demo to a production setup. For teams building assistants that need to handle several modalities at once, this is a solid starting point to evaluate before you wire together multiple single-purpose models.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
transformers
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
MiniCPM-V
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
generative-models
Generative Models by Stability AI
unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities