Feature Extraction
Transformers
Safetensors
Chinese
English
tianmu_emb_uni_adapter_prototype
multimodal
embedding
retrieval
audio
video
image
text
visdoc
qwen3-vl
mmeb-v3
Instructions to use TianmuLab/Tianmu-Emb-Uni with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TianmuLab/Tianmu-Emb-Uni with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="TianmuLab/Tianmu-Emb-Uni")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("TianmuLab/Tianmu-Emb-Uni", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| # Processor Files | |
| This folder stores lightweight tokenizer/processor files copied from the base | |
| models used by the release environment. | |
| - `qwen3_vl_embedding_8b/`: tokenizer and preprocessor files for the | |
| Qwen3-VL-Embedding-8B text/image/video backbone. | |
| - `qwen2_5_omni_7b/`: tokenizer and preprocessor files for the Qwen2.5-Omni-7B | |
| audio processor/audio tower. | |
| The large base model weights are not included in this repository. | |