MTEB GermanLegal1Retrieval
35 models with a score. Leader: Voyage 4 large with 75.5.
| Benchmark profile | |
|---|---|
| Task | Retrieval (de) |
| Metric | nDCG@10 |
| Direction | ↑ higher is better |
| Limitation | Ein Wert gilt fuer die gemessene Modellrevision und die volle Vektorlaenge; gekuerzte Laengen sind nicht gemessen (R-MESS). |
| Source | huggingface.co |
| Licence | CC0 — MTEB-Ergebnisse (embeddings-benchmark/results) |
Ranking (35)
No entry matches the filter.
| Model | Rank | Score | Developer | EU processing: InputEU in | EU processing: OutputEU out | all providers: Inputall in | all providers: Outputall out | Open weights | GPU memory (4-bit, derived)GPU memory |
|---|---|---|---|---|---|---|---|---|---|
| Voyage 4 large | 1 | 75.5 | Voyage AI | – | – | – | – | – no | – |
| most-embed-de | 2 | 75.2 | malteos | – | – | – | – | ✓ yes | from 5 GB |
| Voyage 4 | 3 | 74.8 | Voyage AI | – | – | – | – | – no | – |
| Cohere Embed 4 | 4 | 72.9 | Cohere | – | – | €0.11 | – | – no | – |
| Voyage 4 lite | 5 | 72.6 | Voyage AI | – | – | – | – | – no | – |
| voyage-3.5 | 5 | 72.6 | Voyage AI | – | – | – | – | – no | – |
| gemini-embedding-001 | 7 | 71.5 | – | – | €0.067 | – | – no | – | |
| snowflake-arctic-embed-l-v2.0 | 8 | 69.9 | Snowflake | – | – | – | – | ✓ yes | from 5 GB |
| Qwen3 Embedding 8B | 9 | 69.8 | Alibaba Cloud | €0.009 | – | €0.009 | – | ✓ yes | from 9 GB |
| E5-Mistral-7B | 10 | 69.3 | Microsoft | €0.02 | – | €0.02 | – | ✓ yes | from 8 GB |
| snowflake-arctic-embed-m-v2.0 | 11 | 68.6 | Snowflake | – | – | €0.036 | – | ✓ yes | from 5 GB |
| nomic-embed-text-v2-moe | 12 | 66.9 | Nomic AI | – | – | – | – | ✓ yes | from 5 GB |
| text-embedding-3-large | 13 | 66.7 | OpenAI | €0.13 | – | €0.13 | – | – no | – |
| cohere-embed-multilingual-v3.0 | 14 | 65.2 | Cohere | – | – | €0.09 | – | – no | – |
| multilingual-e5-large-instruct | 15 | 64.1 | Microsoft | €0.03 | €0.02 | €0.009 | – | ✓ yes | from 5 GB |
| Qwen3-Embedding-0.6B | 16 | 61.8 | Alibaba Cloud | – | – | €0.011 | – | ✓ yes | from 5 GB |
| text-embedding-3-small | 16 | 61.8 | OpenAI | €0.02 | – | €0.02 | – | – no | – |
| bekko-embedding-v1-a8m | 18 | 61.1 | hotchpotch | – | – | – | – | ✓ yes | from 5 GB |
| BGE-M3 | 19 | 60.9 | BAAI | €0.0082 | – | €0.0082 | – | ✓ yes | from 5 GB |
| harrier-oss-v1-270m | 20 | 60.5 | Microsoft | – | – | – | – | ✓ yes | from 5 GB |
| bekko-embedding-v1-a25m | 21 | 60.1 | hotchpotch | – | – | – | – | ✓ yes | from 5 GB |
| multilingual-e5-small | 22 | 58.3 | – | – | €0.0067 | – | ✓ yes | from 5 GB | |
| text-embedding-004 | 23 | 57.5 | – | – | – | – | – no | – | |
| gte-multilingual-base | 24 | 56.9 | Alibaba Cloud | – | – | – | – | ✓ yes | from 5 GB |
| text-embedding-005 | 25 | 55.2 | – | – | – | – | – no | – | |
| granite-embedding-311m-multilingual-r2 | 26 | 54.4 | IBM | – | – | – | – | ✓ yes | from 5 GB |
| granite-embedding-97m-multilingual-r2 | 27 | 48.3 | IBM | – | – | – | – | ✓ yes | from 5 GB |
| Cohere Embed 3 Model - English | 28 | 47.25 | Cohere | – | – | €0.09 | – | – | – |
| bge-base-en-v1.5 | 29 | 40.8 | BAAI | – | – | €0.0045 | – | ✓ yes | from 5 GB |
| All MiniLM L12 v2 | 30 | 39.56 | – | – | – | €0.0045 | – | ✓ yes | from 5 GB |
| bge-large-en-v1.5 | 31 | 39.2 | BAAI | – | – | €0.009 | – | ✓ yes | from 5 GB |
| all-mpnet-base-v2 | 32 | 37.96 | – | – | – | €0.0045 | – | ✓ yes | from 5 GB |
| all-MiniLM-L6-v2 | 33 | 37.5 | – | – | – | €0.0045 | – | ✓ yes | from 5 GB |
| mxbai-embed-large-v1 | 34 | 36.9 | Mixedbread | – | – | – | – | ✓ yes | from 5 GB |
| text2vec-base-chinese | 35 | 22.19 | – | – | – | €0.0045 | – | ✓ yes | from 5 GB |
Equal scores share a rank. When filtering, the rank stays the position in the whole list. GPU memory “from …” is a lower bound (parameters × 0.55 GB + 4 GB). Prices: € per 1M tokens. Source: providers' pricing pages, read and checked by searchit. As of 7 Nov 2026.
Scores: MTEB GermanLegal1Retrieval, CC0 — MTEB-Ergebnisse (embeddings-benchmark/results). Taken unchanged from the source.