intfloat / feature-extraction updated 1 year ago

multilingual-e5-large-instruct

Name: multilingual-e5-large-instruct
Author: intfloat

Multilingual E5 Text Embeddings: A Technical Report. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei, arXiv 2024

Params

560 M

Context

512

Downloads 30d

2.3 M

Likes

631

Commercial use: allowed mit Not gated SAFETENSORS 94 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

2.3 M2.1 M

Jul 28Jul 31Aug 3Aug 6

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	f16	1.1 GB	1.8 GB	✅ Runs comfortably

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

Architecture: XLMRobertaModel
Parameters: 560 M
Tensor type: F16
Context length: 512
Vocabulary: 250,002
Layers / heads: 24 / 16
Licence: mit
First seen on the Hub: 2024-02-08
Training datasets: undisclosed
MTEB ArguAna (reported): 31.721
MTEB AmazonPolarityClassification (reported): 96.286809236104
MTEB AmazonReviewsClassification (de) (reported): 52.008299947652
MTEB AmazonReviewsClassification (en) (reported): 55.765103982664
MTEB AmazonReviewsClassification (es) (reported): 48.082345914984
MTEB AmazonReviewsClassification (fr) (reported): 47.68752844642
MTEB AmazonReviewsClassification (ja) (reported): 47.058703766372
MTEB AmazonReviewsClassification (zh) (reported): 43.423719653728
MTEB AmazonCounterfactualClassification (de) (reported): 64.819518175607
MTEB AmazonCounterfactualClassification (en) (reported): 70.048367338627
MTEB AmazonCounterfactualClassification (ja) (reported): 62.572323090776
MTEB AmazonCounterfactualClassification (en-ext) (reported): 61.016273074563
Added to our catalog: 2026-07-28

Compare with

Appears in

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