Ru En RoSBERTa vs Sbert Large MT Nlu Ru
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) Embeddings assign examples to categories, group similar examples, estimate semantic similarity between pairs, classify relationships between pairs, rank relevant documents for queries, find translated sentence pairs, rerank documents according to instructions, assign multiple labels to each example, reorder candidate documents by relevance on MTEB Multilingual v2 · mean over tasks in 1037 languages including 'Auhelawa, Abau, Achi, Achinese, Achuar-Shiwiar, Adzera, Afrikaans, Agarabi; tasks receive equal weight
Ru En RoSBERTa41.6
Sbert Large MT Nlu Ru—
| Benchmark | Ru En RoSBERTa | Sbert Large MT Nlu Ru |
|---|---|---|
| MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) | 41.6 | — |
Specs
| Spec | Ru En RoSBERTa | Sbert Large MT Nlu Ru |
|---|---|---|
| Context | 1K | 1K |
| Weights | Open | Open |
| Released | Jul 2024 | May 2021 |
| Dimensions | 1,024 | 1,024 |
| Max input | 1K tokens | 1K tokens |
| Languages | en, ru | ru |
| Memory (4-bit) | 0.2 GB | 0.3 GB |
| Licence | MIT | — |
| Commercial use | Allowed | Allowed |
What breaks when you switch
Ru En RoSBERTa → Sbert Large MT Nlu Ru
Re-embed your whole corpus›
Vectors from different models can't be mixed.
Languages: loses en
Memory (4-bit): 0.2 GB → 0.3 GB
Sbert Large MT Nlu Ru → Ru En RoSBERTa
Re-embed your whole corpus›
Vectors from different models can't be mixed.