diff --git a/fastembed/text/onnx_embedding.py b/fastembed/text/onnx_embedding.py index 1e9978f7..9691ccc2 100644 --- a/fastembed/text/onnx_embedding.py +++ b/fastembed/text/onnx_embedding.py @@ -180,6 +180,20 @@ sources=ModelSource(hf="jinaai/jina-clip-v1"), model_file="onnx/text_model.onnx", ), + DenseModelDescription( + model="BAAI/bge-m3", + dim=1024, + description=( + "Text embeddings, Unimodal (text), Multilingual (~100 languages), " + "8192 input tokens truncation, " + "Prefixes for queries/documents: not necessary, 2024 year." + ), + license="mit", + size_in_GB=2.27, + sources=ModelSource(hf="BAAI/bge-m3"), + model_file="onnx/model.onnx", + additional_files=["onnx/model.onnx_data"], + ), ] diff --git a/tests/test_text_onnx_embeddings.py b/tests/test_text_onnx_embeddings.py index cdd22d79..a8a1d247 100644 --- a/tests/test_text_onnx_embeddings.py +++ b/tests/test_text_onnx_embeddings.py @@ -68,6 +68,7 @@ "Qdrant/clip-ViT-B-32-text": np.array([0.0083, 0.0103, -0.0138, 0.0199, -0.0069]), "thenlper/gte-base": np.array([0.0038, 0.0355, 0.0181, 0.0092, 0.0654]), "jinaai/jina-clip-v1": np.array([-0.0862, -0.0101, -0.0056, 0.0375, -0.0472]), + "BAAI/bge-m3": np.array([-0.0403941, 0.03703506, -0.02897445, 0.01611726, -0.03569157]), "google/embeddinggemma-300m": np.array( [-0.08181356, 0.0214127, 0.05120273, -0.03690156, -0.0254504] ),