Hugging Face's logo

Urock-AI
/
Eddy-vl_embedding_1.9B_v1

Feature Extraction
Transformers
Safetensors
English
Korean
multilingual
qwen3_vl
vision-language
embedding
multimodal-embedding
mmeb
digital-forensics
custom_code
Model card Files
xet
 Community
Urock-chy commited on
Commit
d5bb69f
·
verified ·
1 Parent(s): 760cac6

Trim README note

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -94,7 +94,7 @@ embedder = VLEmbedder(
94
  # embedder = VLEmbedder(model_id, weights_path="/path/to/model.safetensors", ...)
95
  ```
96
 
97
- > `trust_remote_code=True` is required for `processing_vl.py` (`VLProcessor`) in this repo. Model weights load from `model.safetensors`; no separate Qwen checkpoint download.
98
 
99
  Embeddings are L2-normalized and compared by cosine similarity (dot product). A simple "find the closest image to this text" looks like:
100
 
 
94
  # embedder = VLEmbedder(model_id, weights_path="/path/to/model.safetensors", ...)
95
  ```
96
 
97
+ > `trust_remote_code=True` is required for `processing_vl.py` (`VLProcessor`) in this repo. Model weights load from `model.safetensors`.
98
 
99
  Embeddings are L2-normalized and compared by cosine similarity (dot product). A simple "find the closest image to this text" looks like:
100