netapy's picture
Update README.md
1f42b6f
|
raw
history blame
1.24 kB
---
pipeline_tag: sentence-similarity
tags:
- feature-extraction
- sentence-similarity
license: mit
language:
- fr
- en
---
# Solon Embeddings — base 0.1
SOTA Open source french embedding model.
| Model | Mean Score |
| --- | --- |
| cohere/embed-multilingual-v3 | 0.7402 |
| OrdalieTech/Solon-embeddings-base-0.1 | 0.7306 |
| openai/ada-002 | 0.7290 |
| cohere/embed-multilingual-light-v3 | 0.6945 |
| antoinelouis/biencoder-camembert-base-mmarcoFR | 0.6826 |
| dangvantuan/sentence-camembert-large | 0.6756 |
| voyage/voyage-01 | 0.6753 |
| intfloat/multilingual-e5-large | 0.6660 |
| intfloat/multilingual-e5-base | 0.6597 |
| Sbert/paraphrase-multilingual-mpnet-base-v2 | 0.5975 |
| dangvantuan/sentence-camembert-base | 0.5456 |
| EuropeanParliament/eubert_embedding_v1 | 0.5063 |
These results have been obtained through 9 benchmarks on a variety of text similarity tasks (classification, reranking, STS) :
- AmazonReviewsClassification
- MassiveIntentClassification
- MassiveScenarioClassification
- MTOPDomainClassification
- MTOPIntentClassification
- STS22
- MiraclFRRerank
- OrdalieFRSTS
- OrdalieFRReranking
(evaluation script currently available here : github.com/netapy/mteb)
--------
(Large version comming soon)