384-dimension embeddings
According to the model card, it maps sentences and paragraphs to a 384-dimensional dense vector space, with pooling configured for 384 dimensions and mean-token pooling.
Open Source Model Profile · TaylorAI
bge-micro-v2
bge-micro-v2 is a 17.39M-parameter BERT embedding model from TaylorAI. According to the model card, it produces 384-dimensional vectors distilled from BAAI/bge-small-en-v1.5.
- Publisher
- TaylorAI
- Task
- sentence-similarity
- Model type
- bert
- License
- mit
- Library
- sentence-transformers
- Publication status
- Accepted · not indexed
Model overview
bge-micro-v2 is published by TaylorAI as a bert-based sentence-similarity model. The captured configuration identifies BertModel and Safetensors metadata reports 17389824 parameters. According to the model card, it maps text to 384-dimensional vectors and was distilled from BAAI/bge-small-en-v1.5, with card data recording mit.
Recorded capabilities
BAAI distillation lineage
According to the model card, it was distilled in a 2-step training process from BAAI/bge-small-en-v1.5, with bge-micro as step one.
Sentence-transformers use
The record is tagged with the sentence-transformers library, and the model card documents both SentenceTransformer encode use and direct Transformers use with mean pooling.
512-length transformer block
According to the model card, the SentenceTransformer block lists max_seq_length 512 with a BertModel transformer and mean-token pooling.
MIT licensing
Card data records mit for this repository.
Use cases in the source record
- Semantic search using the documented 384-dimensional sentence and paragraph vectors.
- Clustering workflows that use the publisher-described dense vector mapping for sentences and paragraphs.
Limitations and unknowns
- No independent evaluation scores were extracted; the card only points to the external Sentence Embeddings Benchmark for automated evaluation.
- Provider state is historical snapshot data, not independently refreshed current availability.
- No context-window value beyond the card's 512 max_seq_length field was independently verified.
Source and provenance
Source: TaylorAI/bge-micro-v2
Captured: Unknown. Processed: 2026-09-07T19:34:37.560528+00:00.
bge-micro-v2 This is a sentence-transformers model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search. Distilled in a 2-step training process (bge-micro was step 1) from BAAI/bge-small-en-v1.5 . Usage (Sentence-Transformers) Using this model becomes easy when you have sentence-transformers installed: pip install -U sentence-transformers Then you can use the model like this: from sentence_transformers import SentenceTransformer sentences = [ "This is an example sentence" , "Each sentence is converted" ] model = SentenceTransformer( '{MODEL_NAME}' ) embe…
F001F002F003F004F005F006F007F008F010F011F012F013F014F017