models / embeddings
bge-small-en-v1.5
BAAI/bge-small-en-v1.5
A compact, high-quality English embedding model that punches well above its size. The go-to for cheap, fast retrieval embeddings.
price
$0.008
/ 1M tokens
API providers
0
throughput
193 embeds/s
license
mit
about this model
Overview
BAAI/bge-small-en-v1.5 is a dense embedding model for English text, part of the BGE (BAAI General Embedding) series. It is designed for retrieval tasks such as semantic search and passage ranking. As a version 1.5 model, it features a more reasonable similarity distribution and enhanced retrieval ability without requiring an instruction prefix. For search use cases, the recommended query instruction is: Represent this sentence for searching relevant passages: .
Key Strengths
- Small-scale architecture delivering competitive performance relative to models of similar size.
- Improved similarity distribution and retrieval capability compared to the original v1 release.
- Optimized for low-latency, high-throughput embedding generation.
Best For
- English text embedding for search, document retrieval, and semantic similarity.
- Scenarios where model footprint and inference speed are critical.
Model Specifications
| Property | Value |
|---|---|
| Language | English |
| Model Size | Small |
| Version | v1.5 |
| Task | Embedding / Dense Retrieval |
call it
# OpenAI client - just change base_url from openai import OpenAI client = OpenAI(base_url="https://gigarouter.ai/v1", api_key=KEY) v = client.embeddings.create(model="BAAI/bge-small-en-v1.5", input=["hello world"]) print(v.data[0].embedding[:4])