Sentence Similarity
sentence-transformers
Safetensors
Transformers
qwen3_vl
image-text-to-text
multimodal embedding
qwen
embedding
Instructions to use Qwen/Qwen3-VL-Embedding-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use Qwen/Qwen3-VL-Embedding-2B with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("Qwen/Qwen3-VL-Embedding-2B") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Transformers
How to use Qwen/Qwen3-VL-Embedding-2B with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Qwen/Qwen3-VL-Embedding-2B") model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen3-VL-Embedding-2B") - Notebooks
- Google Colab
- Kaggle
Built a server to make this OpenAI compatible
#21
by sur4guy101 - opened
Hey — getting Qwen3-VL-Embeddings running locally for RAG was kind of a headache (deps, setup, etc).
I put together a small server that makes it OpenAI-compatible so you can just drop it into existing pipelines.
- works with standard OpenAI clients
- handles text + image embeddings
- runs clean locally (tested on a 5070)
If you’re trying to use this model without fighting setup, this might save you some time.
https://github.com/philmcginty/qwen3-vl-embedding-server
Happy to help if anyone gets stuck setting it up.