| library_name: transformers | |
| license: apache-2.0 | |
| language: | |
| - en | |
| tags: | |
| - fill-mask | |
| - masked-lm | |
| - long-context | |
| - modernbert | |
| - mlx | |
| pipeline_tag: fill-mask | |
| inference: false | |
| # mlx-community/ModernBERT-base-4bit | |
| The Model [mlx-community/ModernBERT-base-4bit](https://huggingface.co/mlx-community/ModernBERT-base-4bit) was converted to MLX format from [answerdotai/ModernBERT-base](https://huggingface.co/answerdotai/ModernBERT-base) 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("mlx-community/ModernBERT-base-4bit") | |
| # 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) | |
| ``` | |