Alibaba-NLP/gte-multilingual-base

The gte-multilingual-base model is the latest in the GTE (General Text Embedding) family of models, featuring several key attributes:

Overview

Architecture

ModernBERT

Parameters

305M

Tasks

Encode

Outputs

Dense

Dimensions

Dense: 768

Max Sequence Length

8,192 tokens

License

apache-2.0

Languages

af, ar, az, be, bg, bn, ca, ceb, cs, cy, da, de, el, en, es, et, eu, fa, fi, fr, gl, gu, he, hi, hr, ht, hu, hy, id, is, it, ja, jv, ka, kk, km, kn, ko, ky, lo, lt, lv, mk, ml, mn, mr, ms, my, ne, nl, no, pa, pl, pt, qu, ro, ru, si, sk, sl, so, sq, sr, sv, sw, ta, te, th, tl, tr, uk, ur, vi, yo, zh

View on HuggingFace →

Benchmarks

CQADupstackPhysicsRetrieval

scientific retrieval en

Duplicate question retrieval from StackExchange Physics

Corpus: 38,314 Queries: 1,039

Quality

ndcg at 10 0.4452

Performance L4 b1 c16

Corpus 43.5K tok/s

Corpus p50 48.0ms

Query 4.3K tok/s

Query p50 35.3ms

Reference →

CosQA

technology retrieval en

Code search with natural language queries

Corpus: 6,267 Queries: 500

Quality

map at 10 0.2468

mrr at 10 0.2634

ndcg at 10 0.3251

Performance L4 b1 c16

Corpus 20.4K tok/s

Corpus p50 39.9ms

Query 2.2K tok/s

Query p50 37.2ms

Reference →

FiQA2018

finance retrieval en

Financial opinion mining and question answering

Corpus: 57,599 Queries: 648

Quality

ndcg at 10 0.4482

Performance L4 b1 c16

Corpus 44.2K tok/s

Corpus p50 54.2ms

Query 4.7K tok/s

Query p50 34.9ms

Reference →

LegalBenchConsumerContractsQA

legal retrieval en

Question answering on consumer contracts

Corpus: 153 Queries: 396

Quality

map at 10 0.7011

mrr at 10 0.7032

ndcg at 10 0.7551

Performance L4 b1 c16

Corpus 72.0K tok/s

Corpus p50 114.4ms

Query 5.8K tok/s

Query p50 39.0ms

Reference →

NFCorpus

medical retrieval en

Biomedical literature search from NutritionFacts.org

Corpus: 3,593 Queries: 323

Quality

ndcg at 10 0.3667

map at 10 0.1408

mrr at 10 0.5658

Performance L4 b1 c16

Corpus 60.0K tok/s

Corpus p50 79.5ms

Query 1.4K tok/s

Query p50 49.0ms

Reference →

NanoFiQA2018Retrieval

finance retrieval en

Smaller subset of the FiQA financial QA dataset

Quality

ndcg at 10 0.5990

map at 10 0.5229

mrr at 10 0.6550

Performance L4 b1 c16

Corpus 36.5K tok/s

Corpus p50 63.1ms

Query 3.4K tok/s

Query p50 52.5ms

Reference →

SCIDOCS

scientific retrieval en

Citation prediction, document classification, and recommendation for scientific papers

Corpus: 25,656 Queries: 1,000

Quality

map at 10 0.1076

mrr at 10 0.3208

ndcg at 10 0.1852

Performance L4 b1 c16

Corpus 55.1K tok/s

Corpus p50 56.8ms

Query 4.4K tok/s

Query p50 35.3ms

Reference →

SciFact

scientific retrieval en

Scientific claim verification using research literature

Corpus: 5,183 Queries: 300

Quality

map at 10 0.6951

mrr at 10 0.7046

ndcg at 10 0.7394

Performance L4 b1 c16

Corpus 63.8K tok/s

Corpus p50 69.3ms

Query 6.3K tok/s

Query p50 36.8ms

Reference →

StackOverflowQA

technology retrieval en

Programming question answering from Stack Overflow

Corpus: 19,931 Queries: 1,994

Quality

map at 10 0.8489

mrr at 10 0.8489

ndcg at 10 0.8708

Performance L4 b1 c16

Corpus 55.5K tok/s

Corpus p50 65.9ms

Query 58.5K tok/s

Query p50 75.0ms

Reference →

Overview

Benchmarks

CQADupstackPhysicsRetrieval

CosQA

FiQA2018

LegalBenchConsumerContractsQA

NFCorpus

NanoFiQA2018Retrieval

SCIDOCS

SciFact

StackOverflowQA

Self-hosted inference for search & document processing