Benchmark · Embeddings
MTEB English v2 retrieval
An embedding model's mean nDCG@10 across the 10 retrieval tasks of MTEB(eng, v2), as the MTEB leaderboard's Retrieval column on the English tab.
- Score
- mean nDCG@10 (0 to 100), higher is better
- Run by
- MTEB maintainers (an open community project started at Hugging Face), from results submitted by pull request
- Status
- Current, newest result 20 November 2026
- Updated
- continuous (pull requests to the results repository)
- Data
- the published file, CC0-1.0, read 20 November 2026
Results
12 models we list, each at its best setting, best first.
| # | Model | Score | Range | Who ran it | Setting | Date | |
|---|---|---|---|---|---|---|---|
| 1 | Qwen3-Embedding-8B BB | 69.4 | Submitted by the maker | source | |||
| 2 | Qwen3-Embedding-4B BB | 68.5 | Submitted by the maker | source | |||
| 3 | Gemini Embedding B | 64.3 | Submitted by the maker | model gemini-embedding-001 | source | ||
| 4 | Qwen3-Embedding-0.6B B | 61.8 | Submitted by the maker | source | |||
| 5 | Jina Embeddings and Reranker B | 60.1 | Submitted by the maker | model jina-embeddings-v5-text-small | source | ||
| 5 | jina-embeddings-v5-omni-small D | 60.1 | Submitted by the maker | source | |||
| 7 | OpenAI embeddings BB | 58.0 | Submitted by the maker | model text-embedding-3-large | source | ||
| 8 | Cohere Embed and Rerank BB | 55.7 | Submitted by the maker | model Cohere-embed-english-v3.0 | source | ||
| 8 | EmbeddingGemma 300M D | 55.7 | Submitted by the maker | source | |||
| 10 | Granite Embedding Small English R2 E | 53.9 | Submitted by the maker | source | |||
| 11 | EmbeddingGemma 2 B | 51.3 | Submitted by the maker | source | |||
| 12 | Nomic Embed E | 50.7 | Submitted by the maker | model nomic-embed-text-v1 | source |
Every setting, 20 rows
| Model | As the leaderboard names it | Score | Who ran it | Setting | Date | |
|---|---|---|---|---|---|---|
| Qwen3-Embedding-8B | Qwen/Qwen3-Embedding-8B | 69.4 | Submitted by the maker | source | ||
| Qwen3-Embedding-4B | Qwen/Qwen3-Embedding-4B | 68.5 | Submitted by the maker | source | ||
| Gemini Embedding | google/gemini-embedding-001 | 64.3 | Submitted by the maker | model gemini-embedding-001 | source | |
| Qwen3-Embedding-0.6B | Qwen/Qwen3-Embedding-0.6B | 61.8 | Submitted by the maker | source | ||
| Jina Embeddings and Reranker | jinaai/jina-embeddings-v5-text-small | 60.1 | Submitted by the maker | model jina-embeddings-v5-text-small | source | |
| jina-embeddings-v5-omni-small | jinaai/jina-embeddings-v5-omni-small | 60.1 | Submitted by the maker | source | ||
| Jina Embeddings and Reranker | jinaai/jina-embeddings-v5-omni-nano | 58.8 | Submitted by the maker | model jina-embeddings-v5-omni-nano | source | |
| Jina Embeddings and Reranker | jinaai/jina-embeddings-v5-text-nano | 58.8 | Submitted by the maker | model jina-embeddings-v5-text-nano | source | |
| OpenAI embeddings | openai/text-embedding-3-large | 58.0 | Submitted by the maker | model text-embedding-3-large | source | |
| Jina Embeddings and Reranker | jinaai/jina-embeddings-v4 | 56.1 | Submitted by the maker | model jina-embeddings-v4 | source | |
| Cohere Embed and Rerank | Cohere/Cohere-embed-english-v3.0 | 55.7 | Submitted by the maker | model Cohere-embed-english-v3.0 | source | |
| Cohere Embed and Rerank | Cohere/Cohere-embed-multilingual-v3.0 | 55.7 | Submitted by the maker | model Cohere-embed-multilingual-v3.0 | source | |
| EmbeddingGemma 300M | google/embeddinggemma-300m | 55.7 | Submitted by the maker | source | ||
| Granite Embedding Small English R2 | ibm-granite/granite-embedding-small-english-r2 | 53.9 | Submitted by the maker | source | ||
| OpenAI embeddings | openai/text-embedding-3-small | 53.5 | Submitted by the maker | model text-embedding-3-small | source | |
| Cohere Embed and Rerank | Cohere/Cohere-embed-english-light-v3.0 | 53.2 | Submitted by the maker | model Cohere-embed-english-light-v3.0 | source | |
| Cohere Embed and Rerank | Cohere/Cohere-embed-multilingual-light-v3.0 | 52.1 | Submitted by the maker | model Cohere-embed-multilingual-light-v3.0 | source | |
| EmbeddingGemma 2 | google/embeddinggemma-2 | 51.3 | Submitted by the maker | source | ||
| Nomic Embed | nomic-ai/nomic-embed-text-v1 | 50.7 | Submitted by the maker | model nomic-embed-text-v1 | source | |
| Nomic Embed | nomic-ai/nomic-embed-text-v1.5 | 48.0 | Submitted by the maker | model nomic-embed-text-v1.5 | source |
A rank is among the models we list with a result, not a place on the whole leaderboard, and a model offered by several hosts is shown once, on its own listing. These results don't move our grades. Data: MTEB, embeddings-benchmark/results (CC0), CC0-1.0.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.