LegalQuAD
Models find the German legal document that answers each of 200 questions. The original data also mark the answer text and its starting position.
VAGO's independent leaderboard scores 41 embedding models and a BM25 baseline. It reports nDCG@10 for document retrieval.
- Made by
- Christoph Hoppe and colleagues
- Released
- August 2021
- Kind
- Law
- Level
- University
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Subject
- Law
- Language
- German
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Law and finance
- Last model added
- October 2026
- Scores refreshed
- November 20th, 2026
SauerkrautLM-MoEdernColBERT-115M-Legal-DE tops the leaderboard with 0.87 Qwen3-Embedding-8B leads with 0.70 among models released by October 2026
Best score per entry from VAGO Solutions’s leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen3-Embedding-8B, at 0.70. This view shows only models we track, each at its best.
The newest model with a score here came out in October 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6.1 Terra
- and 16 more
- 0.87 SauerkrautLM-MoEdernColBERT-115M-Legal-DE
- 0.86 topk-embed-v1-small (MTEB) ⚠
- 0.84 topk-embed-v1-xsmall (MTEB)
- 0.83 SauerkrautLM-MoEdernColBERT-27M-Legal-DE
- 0.81 SauerkrautLM-MoEdernColBERT-1.5M-Legal-DE
- 0.78 SauerkrautLM Legal-API
- 0.77 dinghy-law-8b-v1 ⚠
- 0.77 voyage-4-large (API) (MTEB)
- 0.76 Nemotron-3.5-Embed-8B
- 0.76 most-embed-de
- 0.76 Nemotron-3-Embed-8B
- 0.76 BM25
- 0.75 octen-law-8b-v1 ⚠
- 0.74 Nemotron-3-Embed-1B
- 0.74 Nemotron-3.5-Embed-1B
- 0.74 dinghy-law-4b-v1 ⚠
- 0.72 Cohere embed-v4.0 (API) (MTEB)
- 0.71 Euler-Legal-Embedding-V1
- 0.70
Qwen3-Embedding-8B
- 0.70 litil-embed-0.6b ⚠
- 0.69 MoD-Embedding
- 0.69
Qwen3-Embedding-4B
- 0.67 voyage-law-2 (Legal-API) (MTEB)
- 0.66 gemini-embedding-001 (API) (MTEB)
- 0.65 snowflake-arctic-embed-l-v2.0 (MTEB)
- 0.64 jina-embeddings-v5-text-small
- 0.61 bge-multilingual-gemma2
- 0.60 embeddinggemma-300m
- 0.59 jina-embeddings-v3 (MTEB)
- 0.59 greenleaf-law-embed-tiny
- 0.59 dinghy-law-0.6b-v1 ⚠
- 0.58 jina-embeddings-v5-text-nano
- 0.56
Qwen3-Embedding-0.6B
- 0.56 OpenAI text-embedding-3-large (API) (MTEB)
- 0.56
EmbeddingGemma 2
- 0.53 multilingual-e5-large-instruct
- 0.51 bge-m3
- 0.47 multilingual-e5-small
- 0.47 multilingual-e5-base
- 0.37 granite-embedding-278m-multilingual
- 0.33 granite-embedding-107m-multilingual
- 0.13 paraphrase-multilingual-MiniLM-L12-v2
- 1 SauerkrautLM-MoEdernColBERT-115M-Legal-DE 0.87
- 2 topk-embed-v1-small (MTEB) ⚠ 0.86
- 3 topk-embed-v1-xsmall (MTEB) 0.84
- 4 SauerkrautLM-MoEdernColBERT-27M-Legal-DE 0.83
- 5 SauerkrautLM-MoEdernColBERT-1.5M-Legal-DE 0.81
- 6 SauerkrautLM Legal-API 0.78
- 7 dinghy-law-8b-v1 ⚠ st / bf16 / 8192 tokens 0.77
- 8 voyage-4-large (API) (MTEB) 0.77
- 9 Nemotron-3.5-Embed-8B st / bf16 / 8192 tokens 0.76
- 10 most-embed-de st / bf16 / 8192 tokens 0.76
- 11 Nemotron-3-Embed-8B st / bf16 / 8192 tokens 0.76
- 12 BM25 0.76
- 13 octen-law-8b-v1 ⚠ qwen3inst / bf16 / 8192 tokens 0.75
- 14 Nemotron-3-Embed-1B st / bf16 / 8192 tokens 0.74
- 15 Nemotron-3.5-Embed-1B st / bf16 / 8192 tokens 0.74
- 16 dinghy-law-4b-v1 ⚠ qwen3inst / bf16 / 8192 tokens 0.74
- 17 Cohere embed-v4.0 (API) (MTEB) 0.72
- 18 Euler-Legal-Embedding-V1 qwen3inst / bf16 / 8192 tokens 0.71
- 19
Qwen3-Embedding-8B qwen3inst / bf16 / 8192 tokens 0.70
- 20 litil-embed-0.6b ⚠ qwen3inst / bf16 / 8192 tokens 0.70
- 21 MoD-Embedding qwen3inst / bf16 / 8192 tokens 0.69
- 22
Qwen3-Embedding-4B qwen3inst / bf16 / 8192 tokens 0.69
- 23 voyage-law-2 (Legal-API) (MTEB) 0.67
- 24 gemini-embedding-001 (API) (MTEB) 0.66
- 25 snowflake-arctic-embed-l-v2.0 (MTEB) 0.65
- 26 jina-embeddings-v5-text-small st / bf16 / 8192 tokens 0.64
- 27 bge-multilingual-gemma2 gemma2inst / bf16 / 8192 tokens 0.61
- 28 embeddinggemma-300m st / bf16 / 8192 tokens 0.60
- 29 jina-embeddings-v3 (MTEB) 0.59
- 30 greenleaf-law-embed-tiny st / bf16 / 8192 tokens 0.59
- 31 dinghy-law-0.6b-v1 ⚠ qwen3inst / bf16 / 8192 tokens 0.59
- 32 jina-embeddings-v5-text-nano st / bf16 / 8192 tokens 0.58
- 33
Qwen3-Embedding-0.6B qwen3inst / bf16 / 8192 tokens 0.56
- 34 OpenAI text-embedding-3-large (API) (MTEB) 0.56
- 35
EmbeddingGemma 2 st / bf16 / 8192 tokens 0.56
- 36 multilingual-e5-large-instruct e5inst / bf16 / 512 tokens 0.53
- 37 bge-m3 st / bf16 / 8192 tokens 0.51
- 38 multilingual-e5-small e5 / bf16 / 512 tokens 0.47
- 39 multilingual-e5-base e5 / bf16 / 512 tokens 0.47
- 40 granite-embedding-278m-multilingual st / bf16 / 512 tokens 0.37
- 41 granite-embedding-107m-multilingual st / bf16 / 512 tokens 0.33
- 42 paraphrase-multilingual-MiniLM-L12-v2 st / bf16 / 128 tokens 0.13
Source: VAGO Solutions’s leaderboard
Source: VAGO Solutions’s leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | SauerkrautLM-MoEdernColBERT-115M-Legal-DE open weights, 0.1B | 0.87 | huggingface.co | Official leaderboard | ||
| 2 | topk-embed-v1-small (MTEB) ⚠ open weights, 2.2B | 0.86 | huggingface.co | Official leaderboard | ||
| 3 | topk-embed-v1-xsmall (MTEB) open weights, 0.9B | 0.84 | huggingface.co | Official leaderboard | ||
| 4 | SauerkrautLM-MoEdernColBERT-27M-Legal-DE open weights, 0B | 0.83 | huggingface.co | Official leaderboard | ||
| 5 | SauerkrautLM-MoEdernColBERT-1.5M-Legal-DE open weights, 0B | 0.81 | huggingface.co | Official leaderboard | ||
| 6 | SauerkrautLM Legal-API 1.1B | 0.78 | huggingface.co | Official leaderboard | ||
| 7 | dinghy-law-8b-v1 ⚠ open weights, 8B | 0.77 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 8 | voyage-4-large (API) (MTEB) | 0.77 | huggingface.co | Official leaderboard | ||
| 9 | Nemotron-3.5-Embed-8B open weights, 8B | 0.76 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 10 | most-embed-de open weights, 1.1B | 0.76 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 11 | Nemotron-3-Embed-8B open weights, 8B | 0.76 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 12 | BM25 open weights, 0B | 0.76 | huggingface.co | Official leaderboard | ||
| 13 | octen-law-8b-v1 ⚠ open weights, 8B | 0.75 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 14 | Nemotron-3-Embed-1B open weights, 1.1B | 0.74 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 15 | Nemotron-3.5-Embed-1B open weights, 1.1B | 0.74 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 16 | dinghy-law-4b-v1 ⚠ open weights, 4B | 0.74 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 17 | Cohere embed-v4.0 (API) (MTEB) | 0.72 | huggingface.co | Official leaderboard | ||
| 18 | Euler-Legal-Embedding-V1 open weights, 8B | 0.71 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 19 | | 0.70 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 20 | litil-embed-0.6b ⚠ open weights, 0.6B | 0.70 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 21 | MoD-Embedding open weights, 4B | 0.69 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 22 | | 0.69 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 23 | voyage-law-2 (Legal-API) (MTEB) | 0.67 | huggingface.co | Official leaderboard | ||
| 24 | gemini-embedding-001 (API) (MTEB) | 0.66 | huggingface.co | Official leaderboard | ||
| 25 | snowflake-arctic-embed-l-v2.0 (MTEB) open weights, 0.6B | 0.65 | huggingface.co | Official leaderboard | ||
| 26 | jina-embeddings-v5-text-small open weights, 0.7B | 0.64 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 27 | bge-multilingual-gemma2 open weights, 9B | 0.61 | gemma2inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 28 | embeddinggemma-300m open weights, 0.3B | 0.60 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 29 | jina-embeddings-v3 (MTEB) open weights, 0.6B | 0.59 | huggingface.co | Official leaderboard | ||
| 30 | greenleaf-law-embed-tiny open weights, 0.6B | 0.59 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 31 | dinghy-law-0.6b-v1 ⚠ open weights, 0.6B | 0.59 | qwen3inst / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 32 | jina-embeddings-v5-text-nano open weights, 0.2B | 0.58 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 33 | | 0.56 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 34 | OpenAI text-embedding-3-large (API) (MTEB) | 0.56 | huggingface.co | Official leaderboard | ||
| 35 | | 0.56 | st / bf16 / 8192 tokens | Oct 2026 | huggingface.co | Official leaderboard |
| 36 | multilingual-e5-large-instruct open weights, 0.6B | 0.53 | e5inst / bf16 / 512 tokens | huggingface.co | Official leaderboard | |
| 37 | bge-m3 open weights, 0.6B | 0.51 | st / bf16 / 8192 tokens | huggingface.co | Official leaderboard | |
| 38 | multilingual-e5-small open weights, 0.1B | 0.47 | e5 / bf16 / 512 tokens | huggingface.co | Official leaderboard | |
| 39 | multilingual-e5-base open weights, 0.3B | 0.47 | e5 / bf16 / 512 tokens | huggingface.co | Official leaderboard | |
| 40 | granite-embedding-278m-multilingual open weights, 0.3B | 0.37 | st / bf16 / 512 tokens | huggingface.co | Official leaderboard | |
| 41 | granite-embedding-107m-multilingual open weights, 0.1B | 0.33 | st / bf16 / 512 tokens | huggingface.co | Official leaderboard | |
| 42 | paraphrase-multilingual-MiniLM-L12-v2 open weights, 0.1B | 0.13 | st / bf16 / 128 tokens | huggingface.co | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 0.70 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 2 | | 0.69 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 3 | | 0.56 | qwen3inst / bf16 / 8192 tokens | Jun 2025 | huggingface.co | Official leaderboard |
| 4 | | 0.56 | st / bf16 / 8192 tokens | Oct 2026 | huggingface.co | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: VAGO Solutions’s leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: VAGO Solutions’s leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (42 entries).
- Pass Models scored 4 models we track, plus 38 other entries on its leaderboard.
- Pass Newest score The newest is from October 2026, for EmbeddingGemma 2.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on VAGO Solutions: 4 rows for models we track and 38 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.