MODELS
BAAI/bge-base-en-v1.5
by BAAI
Overview
English BGE v1.5 base embedding model for feature extraction, search, and RAG retrieval workflows.
Details
BAAI/bge-base-en-v1.5 is listed on Hugging Face for feature extraction. BAAI’s BGE documentation lists it under BGE-v1.5 as an English model with 109M parameters and a 438 MB size, and provides FlagModel inference usage. The official FlagEmbedding repository describes BGE as a one-stop retrieval toolkit for search and RAG, with quick-start examples that load BAAI/bge-base-en-v1.5 via FlagAutoModel.from_finetuned.
When to Use
Use for English embedding or feature-extraction tasks where the BGE v1.5 base model size is appropriate. Use in search or RAG retrieval pipelines built around the FlagEmbedding toolkit. Evaluate when you need a documented BGE-v1.5 English model with published parameter and file-size metadata.
Getting Started
- Open the Hugging Face model page at https://huggingface.co/BAAI/bge-base-en-v1.5 to review the model entry.
- Read the BGE v1/v1.5 documentation for the listed model metadata and FlagModel inference example.
- Install or review the official FlagEmbedding repository and follow its quick start that loads BAAI/bge-base-en-v1.5 via FlagAutoModel.from_finetuned.
- Run a small embedding or retrieval test before using it in a production search or RAG workflow.
Key Features
- •BGE-v1.5 English base embedding model
- •109M parameters according to the BGE documentation
- •438 MB model size according to the BGE documentation
- •Supported by FlagEmbedding examples and quick-start loading via FlagAutoModel.from_finetuned
- •Hugging Face task listing indicates feature extraction
Capabilities
- •feature-extraction
- •embeddings
- •retrieval
- •search
- •RAG
Last updated Jun 4, 2026