SkMTEB - Bitext Mining: leaderboard
SkMTEB is the Slovak Massive Text Embedding Benchmark: 31 datasets across seven task types (bitext mining, classification, clustering, pair classification, reranking, retrieval, semantic textual similarity). This board reports Mean F1 (%) over the six bitext-mining datasets; higher is better.
Metric: Mean F1 (%) over the six bitext-mining datasets; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 28 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LaBSE | 97.48 | |
| 2 | multilingual-e5-large-instruct | 97.09 | |
| 3 | text-embedding-3-large | 96.79 | |
| 4 | gemini-embedding-001 | 96.76 | |
| 5 | jina-embeddings-v3 | 96.43 | |
| 6 | multilingual-e5-large | 96.29 | |
| 7 | bge-m3 | 96.29 | |
| 8 | paraphrase-multilingual-mpnet-base-v2 | 96.23 | |
| 9 | paraphrase-multilingual-MiniLM-L12-v2 | 94.91 | |
| 10 | multilingual-e5-base | 94.76 |
Interactive version: theaggregate.ai/benchmark?slug=skmteb-bitext-mining · How It Works · Data refreshed daily, snapshot 2026-11-13.