AI Tools.

Search

sentence similarity by TaylorAI

bge-micro-v2

BGE-Micro-v2 is a heavily distilled BERT embedding model targeting near-zero latency sentence encoding with acceptable MTEB scores. Extremely small footprint allows embedding generation in CPU-only or mobile environments. MIT-licensed with ONNX and transformers.js support.

Summary text generated by an automated pipeline from the model card · Not individually reviewed or run by us · How this page is made

From the model card

Fields below are copied from the tags and counters on the HuggingFace repository TaylorAI/bge-micro-v2 at our last fetch. They are set by the uploader, not verified by us; rows with no tag are omitted. How this page is made.

Publisher (HF namespace)
TaylorAI
Pipeline tag
sentence-similarity
Library
Sentence Transformers, Transformers
Framework tags
PyTorch
Weight formats
ONNX, safetensors
License tag
mit — read the license file in the repo before relying on it
Downloads (HF counter at last fetch)
932,343
Likes (HF counter at last fetch)
65
Model card
https://huggingface.co/TaylorAI/bge-micro-v2

Use cases

  • CPU-only embedding generation for resource-constrained services
  • Mobile semantic search with on-device embedding
  • High-throughput embedding where latency is the primary constraint
  • Browser-side semantic similarity via transformers.js/ONNX

Pros

  • MIT license
  • ONNX and sentence-transformers compatible for broad deployment
  • Tiny size — very fast CPU inference
  • text-embeddings-inference compatible

Cons

  • Micro scale sacrifices significant embedding quality vs bge-small-en-v1.5
  • English-only
  • 512 token input limit
  • MTEB scores trail all larger BGE variants — not suitable for quality-sensitive retrieval

Tags

sentence-transformerspytorchonnxsafetensorsbertfeature-extractionsentence-similaritytransformersmteblicense:mitmodel-indextext-embeddings-inferenceendpoints_compatibleregion:usdeploy:azure