BAAI / feature-extraction updated 2 years ago

bge-base-en-v1.5

If you are looking for a model that supports more languages, longer texts, and other retrieval methods, you can try using bge-m3.

Params

110 M

Context

512

Downloads 30d

10.0 M

Likes

462

Commercial use: allowed mit Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

10.0 M8.7 M

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f32	0.4 GB	1.0 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture: BertModel
Parameters: 110 M
Tensor type: F32
Context length: 512
Vocabulary: 30,522
Layers / heads: 12 / 12
Licence: mit
First seen on the Hub: 2023-09-11
Training datasets: undisclosed
MTEB ArguAna (reported): 75.747
MTEB BIOSSES (reported): 88.19658793234
MTEB ArxivClusteringP2P (reported): 48.748845396794
MTEB ArxivClusteringS2S (reported): 42.807589381072
MTEB BiorxivClusteringP2P (reported): 39.102637629289
MTEB BiorxivClusteringS2S (reported): 36.697115174267
MTEB AskUbuntuDupQuestions (reported): 74.280657784813
MTEB Banking77Classification (reported): 86.925153579737
MTEB CQADupstackAndroidRetrieval (reported): 32.327
MTEB AmazonPolarityClassification (reported): 93.377410300062
MTEB AmazonReviewsClassification (en) (reported): 48.146462697783
MTEB AmazonCounterfactualClassification (en) (reported): 70.169022526114
Added to our catalog: 2026-07-28

Family

Base model and the most-downloaded derivatives in the catalog.

Derivative bge-base-en-v1.5 Xenova 2.2 M Derivative bge-base-en-v1.5-course-recommender-v5 datasocietyco 1.8 M

Compare with

bge-small-en-v1.5 72.4 M bge-large-en-v1.5 13.1 M Qwen3-Embedding-0.6B 9.8 M all-MiniLM-L6-v2 2.9 M bge-base-en-v1.5 2.2 M bge-large-zh-v1.5 1.8 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB