Qwen3-Embedding-0.6B
Qwen3-Embedding-0.6B places in the top three on 7 benchmarks.
45
Benchmarks with a score
0
Ranked first, out of five or more models
7
In the top three
Qwen3-Embedding-0.6B
Open weights Small
No Epoch Capabilities Index score yet
- CLICK-ID 64.8% #2 /5
- Estonian Valence Corpus 41.1% #4 /5
- Greek Legal Code 38.8% #3 /5
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CLICK-ID | 64.8% | #2 of 5 | Official leaderboard | |
| Ask Ubuntu duplicate question retrieval | 0.65 | #3 of 5 | Official leaderboard | |
| DBpedia 14 | 98.8% | #3 of 5 | Official leaderboard | |
| Amazon Counterfactual Detection | 90.9% | #3 of 4 | Official leaderboard | |
| CTKFacts | 0.83 | #3 of 4 | Official leaderboard | |
| Faroese Semantic Textual Similarity | 0.74 | #3 of 4 | Official leaderboard | |
| FollowIR | 0.05 | #3 of 4 | Official leaderboard | |
| IndicLID benchmark | 93.1% | #3 of 4 | Official leaderboard | |
| JSICK | 86.6% | #3 of 4 | Official leaderboard | |
| OBLIQ-Bench | 0.05 | #3 of 4 | Official leaderboard | |
| ScaLA | 50.5% | #3 of 4 | Official leaderboard | |
| SemEval 2022 Task 8 multilingual news article similarity | 0.71 | #3 of 4 | Official leaderboard | |
| Estonian Valence Corpus | 41.1% | #4 of 5 | Official leaderboard | |
| IN22-Gen | 75.8% | #4 of 5 | Official leaderboard | |
| Tatoeba sentence retrieval | 58% | #4 of 5 | Official leaderboard | |
| Norwegian Courts Bitext Mining | 92.7% | #6 of 6 | Official leaderboard | |
| DiaBLa | 0.81 | #5 of 5 | Official leaderboard | |
| Corpus for Emotion Detection in Russian | 49.9% | #4 of 4 | Official leaderboard | |
| Korean Sarcasm Dataset | 55.9% | #4 of 4 | Official leaderboard | |
| MTEB-BR | 0.62 | #4 of 4 | Official leaderboard | |
| SemEval 2024 Task 1 semantic textual relatedness | 0.59 | #4 of 4 | Official leaderboard | |
| BRIGHT-Pro | 29.9 | #3 of 3 | Official leaderboard | |
| CLIMATE-FEVER | 0.42 | #3 of 3 | Official leaderboard | |
| FarSick | 0.67 | #3 of 3 | Official leaderboard | |
| Q2D-Web | 71.2% | #3 of 3 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CodeSearchNet Challenge | 0.91 | #3 of 4 | Official leaderboard | |
| CodeTransOcean | 31.4 | #4 of 4 | Official leaderboard | |
| CoSQA | 0.36 | #4 of 4 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AILA Statute Retrieval | 0.79 | #3 of 5 | Official leaderboard | |
| BEIR FiQA-2018 Retrieval | 0.47 | #3 of 5 | Official leaderboard | |
| Financial PhraseBank | 90.6% | #3 of 5 | Official leaderboard | |
| Greek Legal Code | 38.8% | #3 of 5 | Official leaderboard | |
| Bank Question Corpus | 0.65 | #3 of 4 | Official leaderboard | |
| GerLayQA | 0.16 | #3 of 4 | Official leaderboard | |
| LegalQuAD | 0.56 | #3 of 4 | Official leaderboard | |
| MultiEURLEX | 0.05 | #3 of 4 | Official leaderboard | |
| NFCorpus | 0.37 | #3 of 4 | Official leaderboard | |
| BANKING77 | 81% | #4 of 5 | Official leaderboard | |
| GerDaLIR Small | 0.36 | #4 of 4 | Official leaderboard | |
| GerLeRB | 0.48 | #4 of 4 | Official leaderboard | |
| GreekBarRetrieval | 0.14 | #4 of 4 | Official leaderboard | |
| Italian-Administrative-CaseHold | 61.7% | #4 of 4 | Official leaderboard | |
| OpenLegalRetrieval-DE | 0.32 | #4 of 4 | Official leaderboard | |
| Malt Freelancer Project Matching | 50.4% | #1 of 1 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MTEB | 67% | #3 of 4 | Official leaderboard |
Who measured it?
Independent runs cover all 45 of its benchmarks.
- An independent run only 45
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards