UAE Large V1
by WhereIsAI
Open source · 1M downloads · 237 likes
3.0
(237 reviews)EmbeddingAPI & LocalAbout
UAE Large V1 is a universal text embedding model designed to generate precise and versatile vector representations of English sentences. It excels particularly in tasks requiring fine-grained semantic understanding, such as text similarity or information retrieval, thanks to an angle-based optimization that enhances discrimination between embeddings. The model stands out for its state-of-the-art performance on recognized benchmarks, notably achieving an average score of 64.64 on the MTEB leaderboard, making it a reliable tool for a wide range of applications. Its use cases include similarity analysis, text classification, and integration into advanced search systems. Unlike more specialized models, UAE Large V1 offers remarkable flexibility, suitable for both academic and industrial needs.
Documentation
Universal AnglE Embedding
📢 WhereIsAI/UAE-Large-V1 is licensed under MIT. Feel free to use it in any scenario.
If you use it for academic papers, you could cite us via 👉 citation info.
🤝 Follow us on:
- GitHub: https://github.com/SeanLee97/AnglE.
- Preprint Paper: AnglE-optimized Text Embeddings
- Conference Paper: AoE: Angle-optimized Embeddings for Semantic Textual Similarity (ACL24)
- 📘 Documentation: https://angle.readthedocs.io/en/latest/index.html
Welcome to using AnglE to train and infer powerful sentence embeddings.
🏆 Achievements
- 📅 May 16, 2024 | AnglE's paper is accepted by ACL 2024 Main Conference
- 📅 Dec 4, 2023 | 🔥 Our universal English sentence embedding
WhereIsAI/UAE-Large-V1achieves SOTA on the MTEB Leaderboard with an average score of 64.64!
🧑🤝🧑 Siblings:
- WhereIsAI/UAE-Code-Large-V1: This model can be used for code or GitHub issue similarity measurement.
Usage
1. angle_emb
Bash
python -m pip install -U angle-emb
- Non-Retrieval Tasks
There is no need to specify any prompts.
Python
from angle_emb import AnglE
from angle_emb.utils import cosine_similarity
angle = AnglE.from_pretrained('WhereIsAI/UAE-Large-V1', pooling_strategy='cls').cuda()
doc_vecs = angle.encode([
'The weather is great!',
'The weather is very good!',
'i am going to bed'
], normalize_embedding=True)
for i, dv1 in enumerate(doc_vecs):
for dv2 in doc_vecs[i+1:]:
print(cosine_similarity(dv1, dv2))
- Retrieval Tasks
For retrieval purposes, please use the prompt Prompts.C for query (not for document).
Python
from angle_emb import AnglE, Prompts
from angle_emb.utils import cosine_similarity
angle = AnglE.from_pretrained('WhereIsAI/UAE-Large-V1', pooling_strategy='cls').cuda()
qv = angle.encode(Prompts.C.format(text='what is the weather?'))
doc_vecs = angle.encode([
'The weather is great!',
'it is rainy today.',
'i am going to bed'
])
for dv in doc_vecs:
print(cosine_similarity(qv[0], dv))
2. sentence transformer
Python
from angle_emb import Prompts
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("WhereIsAI/UAE-Large-V1").cuda()
qv = model.encode(Prompts.C.format(text='what is the weather?'))
doc_vecs = model.encode([
'The weather is great!',
'it is rainy today.',
'i am going to bed'
])
for dv in doc_vecs:
print(1 - spatial.distance.cosine(qv, dv))
3. Infinity
Infinity is a MIT licensed server for OpenAI-compatible deployment.
Bash
docker run --gpus all -v $PWD/data:/app/.cache -p "7997":"7997" \
michaelf34/infinity:latest \
v2 --model-id WhereIsAI/UAE-Large-V1 --revision "369c368f70f16a613f19f5598d4f12d9f44235d4" --dtype float16 --batch-size 32 --device cuda --engine torch --port 7997
Citation
If you use our pre-trained models, welcome to support us by citing our work:
INI
@article{li2023angle,
title={AnglE-optimized Text Embeddings},
author={Li, Xianming and Li, Jing},
journal={arXiv preprint arXiv:2309.12871},
year={2023}
}
Capabilities & Tags
sentence-transformersonnxsafetensorsopenvinobertfeature-extractionmtebsentence_embeddingfeature_extractiontransformers
Links & Resources