| library_name: transformers | |
| tags: | |
| - colpali | |
| - mlx | |
| license: apache-2.0 | |
| datasets: | |
| - vidore/colpali_train_set | |
| language: | |
| - en | |
| base_model: | |
| - vidore/colqwen2-base | |
| pipeline_tag: visual-document-retrieval | |
| # thoddnn/colqwen2-v1.0-mlx | |
| The Model [thoddnn/colqwen2-v1.0-mlx](https://huggingface.co/thoddnn/colqwen2-v1.0-mlx) was converted to MLX format from [vidore/colqwen2-v1.0-hf](https://huggingface.co/vidore/colqwen2-v1.0-hf) using mlx-lm version **0.0.3**. | |
| ## Use with mlx | |
| ```bash | |
| pip install mlx-embeddings | |
| ``` | |
| ```python | |
| from mlx_embeddings import load, generate | |
| import mlx.core as mx | |
| model, tokenizer = load("thoddnn/colqwen2-v1.0-mlx") | |
| # For text embeddings | |
| output = generate(model, processor, texts=["I like grapes", "I like fruits"]) | |
| embeddings = output.text_embeds # Normalized embeddings | |
| # Compute dot product between normalized embeddings | |
| similarity_matrix = mx.matmul(embeddings, embeddings.T) | |
| print("Similarity matrix between texts:") | |
| print(similarity_matrix) | |
| ``` | |