Italian-Administrative-CaseHold
Models classify Italian administrative judgments paired with their official holdings by legal subject.
MTEB scores the task using accuracy.
- Made by
- Daniele Licari and colleagues
- Kind
- Law
- Level
- University
- Scored
- Checked automatically
- Answers
- Prediction
- Tasks
- All public
- Measures
- Capability
- Subject
- Law
- Language
- Italian
- 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
bigscience/sgpt-bloom-7b1-msmarco tops the leaderboard with 94.4% EmbeddingGemma 2 leads with 68.9% among models released by October 2026
Best score per entry from the benchmark's own 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 EmbeddingGemma 2, at 68.9%. 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
- 94.4 bigscience/sgpt-bloom-7b1-msmarco
- 92.1 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised
- 91.7 WhereIsAI/UAE-Large-V1
- 91.0 jinaai/jina-embeddings-v5-omni-small
- 91.0 jinaai/jina-embeddings-v5-text-small
- 89.8 mixedbread-ai/mxbai-embed-large-v1
- 88.5 izhx/udever-bloom-7b1
- 87.4 jinaai/jina-embeddings-v5-omni-nano
- 87.4 jinaai/jina-embeddings-v5-text-nano
- 85.5 BidirLM/BidirLM-Omni-2.5B-Embedding
- 85.3 Bytedance/Seed1.6-embedding-1215
- 85.0 sdadas/mmlw-e5-large
- 85.0 tencent/KaLM-Embedding-Gemma3-12B-2511
- 85.0 sdadas/mmlw-e5-base
- 84.4 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse
- 84.2 sdadas/mmlw-roberta-base
- 84.0 sdadas/mmlw-roberta-large
- 82.5 Mihaiii/Ivysaur
- 81.6 sdadas/mmlw-e5-small
- 80.3 avsolatorio/NoInstruct-small-Embedding-v0
- 79.3 izhx/udever-bloom-3b
- 79.3 Lajavaness/bilingual-embedding-small
- 79.2 microsoft/harrier-oss-v1-27b
- 78.8 microsoft/harrier-oss-v1-0.6b
- 78.7 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka
- 78.0 izhx/udever-bloom-1b1
- 78.0 BidirLM/BidirLM-1.7B-Embedding
- 76.6 microsoft/harrier-oss-v1-270m
- 76.2 BidirLM/BidirLM-0.6B-Embedding
- 76.0 Haon-Chen/speed-embedding-7b-instruct
- 75.9 BAAI/bge-large-en-v1.5
- 75.1 GritLM/GritLM-8x7B
- 74.6 avsolatorio/GIST-large-Embedding-v0
- 73.5 GritLM/GritLM-7B
- 73.3 google/gemini-embedding-001
- 73.2 nvidia/llama-embed-nemotron-8b
- 73.0 codefuse-ai/F2LLM-v2-14B
- 72.8 Cohere/Cohere-embed-multilingual-light-v3.0
- 72.5 thenlper/gte-large
- 72.4 Mira190/Euler-Legal-Embedding-V1
- 72.4 Alibaba-NLP/gte-Qwen1.5-7B-instruct
- 72.2 codefuse-ai/F2LLM-v2-8B
- 72.2 ICT-TIME-and-Querit/BOOM_4B_v1
- 72.0 ibm-granite/granite-embedding-278m-multilingual
- 72.0 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826
- 71.8 sentence-transformers/paraphrase-multilingual-mpnet-base-v2
- 71.6 Octen/Octen-Embedding-8B
- 71.6 Alibaba-NLP/gte-Qwen2-7B-instruct
- 71.6 NovaSearch/stella_en_1.5B_v5
- 71.5 codefuse-ai/F2LLM-v2-4B
- 71.4 Snowflake/snowflake-arctic-embed-m-v2.0
- 71.3 Salesforce/SFR-Embedding-Mistral
- 71.0 Linq-AI-Research/Linq-Embed-Mistral
- 70.9 BidirLM/BidirLM-1B-Embedding
- 70.7 infgrad/stella-base-en-v2
- 70.5 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1
- 70.4 google/embeddinggemma-300m
- 70.1 openai/text-embedding-3-large
- 70.1 Alibaba-NLP/gte-Qwen2-1.5B-instruct
- 69.9 manu/sentence_croissant_alpha_v0.2
- 69.7 BAAI/bge-base-en-v1.5
- 69.7 thenlper/gte-base
- 69.6 OrdalieTech/Solon-embeddings-large-0.1
- 69.4 ai-sage/Giga-Embeddings-instruct-3B-0826
- 69.4 nomic-ai/nomic-embed-text-v2-moe
- 69.3 intfloat/e5-base-v2
- 68.9
EmbeddingGemma 2
- 68.8 BAAI/bge-small-en-v1.5
- 68.8 NovaSearch/stella_en_400M_v5
- 68.4 Lajavaness/bilingual-embedding-large
- 68.3 mongodb/voyage-3-lite
- 68.3 Mihaiii/gte-micro-v4
- 68.1 Snowflake/snowflake-arctic-embed-l-v2.0
- 68.0 Cohere/Cohere-embed-multilingual-v3.0
- 68.0 NovaSearch/jasper_en_vision_language_v1
- 67.9 codefuse-ai/F2LLM-v2-1.7B
- 67.9 intfloat/e5-mistral-7b-instruct
- 67.8 abhinand/MedEmbed-small-v0.1
- 67.7 BAAI/bge-m3
- 67.6 HIT-TMG/KaLM-embedding-multilingual-mini-v1
- 67.2 sentence-transformers/all-MiniLM-L12-v2
- 67.0 Snowflake/snowflake-arctic-embed-m-v1.5
- 67.0 Cohere/Cohere-embed-english-v3.0
- 66.8 thenlper/gte-small
- 66.8 intfloat/multilingual-e5-large
- 66.7
Qwen3-Embedding-4B
- 66.5 mongodb/voyage-3
- 66.5 Snowflake/snowflake-arctic-embed-m
- 66.3 hotchpotch/bekko-embedding-v1-a25m
- 66.2 Snowflake/snowflake-arctic-embed-l
- 66.2 jinaai/jina-embedding-s-en-v1
- 66.2 hotchpotch/bekko-embedding-v1-a8m
- 66.1
Qwen3-Embedding-8B
- 65.9 intfloat/multilingual-e5-small
- 65.7 manu/sentence_croissant_alpha_v0.4
- 65.5 avsolatorio/GIST-Embedding-v0
- 65.5 manu/bge-m3-custom-fr
- 65.4 izhx/udever-bloom-560m
- 65.3 albertobarnabo/bge-m3-italian
- 65.3 zeroentropy/zembed-1
- 65.1 intfloat/e5-small-v2
- 64.9 deepvk/USER-bge-m3
- 64.8 Lajavaness/bilingual-embedding-base
- 64.7 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet
- 64.5 codefuse-ai/F2LLM-v2-330M
- 64.3 dwzhu/e5-base-4k
- 64.2 Snowflake/snowflake-arctic-embed-s
- 64.1 Salesforce/SFR-Embedding-2_R
- 64.0 intfloat/e5-large-v2
- 63.6 intfloat/e5-base
- 63.4 ibm-granite/granite-embedding-107m-multilingual
- 63.0 BidirLM/BidirLM-270M-Embedding
- 63.0 openai/text-embedding-3-small
- 62.9 intfloat/multilingual-e5-large-instruct
- 62.9 nomic-ai/nomic-embed-text-v1-ablated
- 62.9 Alibaba-NLP/gte-multilingual-base
- 62.9 nomic-ai/nomic-embed-text-v1
- 62.8 jinaai/jina-embeddings-v3
- 62.4 Mihaiii/Bulbasaur
- 62.4 nomic-ai/nomic-embed-text-v1-unsupervised
- 62.3 deepfile/embedder-100p
- 62.2 codefuse-ai/F2LLM-v2-0.6B
- 62.1 mongodb/voyage-3.5
- 62.0 codefuse-ai/F2LLM-v2-160M
- 62.0 manu/sentence_croissant_alpha_v0.3
- 61.7
Qwen3-Embedding-0.6B
- 61.5 intfloat/multilingual-e5-base
- 61.4 Snowflake/snowflake-arctic-embed-xs
- 61.3 avsolatorio/GIST-small-Embedding-v0
- 61.2 google/text-multilingual-embedding-002
- 60.7 avsolatorio/GIST-all-MiniLM-L6-v2
- 60.5 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1
- 60.4 Cohere/Cohere-embed-english-light-v3.0
- 59.2 sentence-transformers/all-mpnet-base-v2
- 58.9 hotchpotch/bekko-static-embedding-v1
- 58.6 ibm-granite/granite-embedding-30m-english
- 58.4 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet
- 58.0 codefuse-ai/F2LLM-v2-80M
- 56.6 Mihaiii/Venusaur
- 56.4 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka
- 55.7 ibm-granite/granite-embedding-97m-multilingual-r2
- 55.4 ibm-granite/granite-embedding-311m-multilingual-r2
- 55.4 nomic-ai/nomic-embed-text-v1.5
- 55.3 Gameselo/STS-multilingual-mpnet-base-v2
- 55.3 Mihaiii/Wartortle
- 54.6 sentence-transformers/static-similarity-mrl-multilingual-v1
- 54.5 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
- 54.3 intfloat/e5-large
- 53.7 jinaai/jina-embedding-b-en-v1
- 53.4 nvidia/NV-Embed-v1
- 53.3 sentence-transformers/LaBSE
- 53.2 omarelshehy/arabic-english-sts-matryoshka
- 53.1 Jaume/gemma-2b-embeddings
- 52.4 sergeyzh/LaBSE-ru-turbo
- 51.9 minishlab/potion-multilingual-128M
- 51.2 brahmairesearch/slx-v0.1
- 51.2 sentence-transformers/all-MiniLM-L6-v2
- 50.9 shibing624/text2vec-base-multilingual
- 50.7 cnmoro/static-nomic-384-pten-v2-st
- 50.4 intfloat/e5-small
- 50.1 deepvk/USER-base
- 49.4 ibm-granite/granite-embedding-125m-english
- 48.4 Mihaiii/Squirtle
- 48.0 consciousAI/cai-lunaris-text-embeddings
- 45.9 Mihaiii/gte-micro
- 45.4 ai-forever/ru-en-RoSBERTa
- 44.6 aari1995/German_Semantic_STS_V2
- 43.7 nvidia/NV-Embed-v2
- 42.3 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka
- 41.6 Snowflake/snowflake-arctic-embed-m-long
- 40.9 cointegrated/LaBSE-en-ru
- 40.7 deepvk/deberta-v1-base
- 40.7 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka
- 40.0 minishlab/M2V_base_glove_subword
- 36.3 DeepPavlov/distilrubert-small-cased-conversational
- 34.9 minishlab/M2V_base_output
- 34.7 DeepPavlov/rubert-base-cased
- 33.2 silma-ai/silma-embeddding-matryoshka-v0.1
- 32.3 minishlab/M2V_base_glove
- 31.6 minishlab/potion-base-8M
- 30.5 DeepPavlov/rubert-base-cased-sentence
- 30.4 cointegrated/rubert-tiny
- 27.4 minishlab/potion-base-4M
- 23.3 malenia1/ternary-weight-embedding
- 23.1 ai-forever/sbert_large_nlu_ru
- 22.9 minishlab/potion-base-2M
- 22.9 ai-forever/sbert_large_mt_nlu_ru
- 18.2 cointegrated/rubert-tiny2
- 18.1 sergeyzh/rubert-tiny-turbo
- 16.6 jinaai/jina-embeddings-v2-small-en
- 7.33 jinaai/jina-embeddings-v2-base-en
- 2.35 mteb/baseline-random-encoder
- 1 bigscience/sgpt-bloom-7b1-msmarco 94.4%
- 2 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised 92.1%
- 3 WhereIsAI/UAE-Large-V1 91.7%
- 4 jinaai/jina-embeddings-v5-omni-small 91%
- 5 jinaai/jina-embeddings-v5-text-small 91%
- 6 mixedbread-ai/mxbai-embed-large-v1 89.8%
- 7 izhx/udever-bloom-7b1 88.5%
- 8 jinaai/jina-embeddings-v5-omni-nano 87.4%
- 9 jinaai/jina-embeddings-v5-text-nano 87.4%
- 10 BidirLM/BidirLM-Omni-2.5B-Embedding 85.5%
- 11 Bytedance/Seed1.6-embedding-1215 85.3%
- 12 sdadas/mmlw-e5-large 85%
- 13 tencent/KaLM-Embedding-Gemma3-12B-2511 85%
- 14 sdadas/mmlw-e5-base 85%
- 15 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse 84.4%
- 16 sdadas/mmlw-roberta-base 84.2%
- 17 sdadas/mmlw-roberta-large 84%
- 18 Mihaiii/Ivysaur 82.5%
- 19 sdadas/mmlw-e5-small 81.6%
- 20 avsolatorio/NoInstruct-small-Embedding-v0 80.3%
- 21 izhx/udever-bloom-3b 79.3%
- 22 Lajavaness/bilingual-embedding-small 79.3%
- 23 microsoft/harrier-oss-v1-27b 79.2%
- 24 microsoft/harrier-oss-v1-0.6b 78.8%
- 25 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka 78.7%
- 26 izhx/udever-bloom-1b1 78%
- 27 BidirLM/BidirLM-1.7B-Embedding 78%
- 28 microsoft/harrier-oss-v1-270m 76.6%
- 29 BidirLM/BidirLM-0.6B-Embedding 76.2%
- 30 Haon-Chen/speed-embedding-7b-instruct 76%
- 31 BAAI/bge-large-en-v1.5 75.9%
- 32 GritLM/GritLM-8x7B 75.1%
- 33 avsolatorio/GIST-large-Embedding-v0 74.6%
- 34 GritLM/GritLM-7B 73.5%
- 35 google/gemini-embedding-001 73.3%
- 36 nvidia/llama-embed-nemotron-8b 73.2%
- 37 codefuse-ai/F2LLM-v2-14B 73%
- 38 Cohere/Cohere-embed-multilingual-light-v3.0 72.8%
- 39 thenlper/gte-large 72.5%
- 40 Mira190/Euler-Legal-Embedding-V1 72.4%
- 41 Alibaba-NLP/gte-Qwen1.5-7B-instruct 72.4%
- 42 codefuse-ai/F2LLM-v2-8B 72.2%
- 43 ICT-TIME-and-Querit/BOOM_4B_v1 72.2%
- 44 ibm-granite/granite-embedding-278m-multilingual 72%
- 45 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 72%
- 46 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 71.8%
- 47 Octen/Octen-Embedding-8B 71.6%
- 48 Alibaba-NLP/gte-Qwen2-7B-instruct 71.6%
- 49 NovaSearch/stella_en_1.5B_v5 71.6%
- 50 codefuse-ai/F2LLM-v2-4B 71.5%
- 51 Snowflake/snowflake-arctic-embed-m-v2.0 71.4%
- 52 Salesforce/SFR-Embedding-Mistral 71.3%
- 53 Linq-AI-Research/Linq-Embed-Mistral 71%
- 54 BidirLM/BidirLM-1B-Embedding 70.9%
- 55 infgrad/stella-base-en-v2 70.7%
- 56 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 70.5%
- 57 google/embeddinggemma-300m 70.4%
- 58 openai/text-embedding-3-large 70.1%
- 59 Alibaba-NLP/gte-Qwen2-1.5B-instruct 70.1%
- 60 manu/sentence_croissant_alpha_v0.2 69.9%
- 61 BAAI/bge-base-en-v1.5 69.7%
- 62 thenlper/gte-base 69.7%
- 63 OrdalieTech/Solon-embeddings-large-0.1 69.6%
- 64 ai-sage/Giga-Embeddings-instruct-3B-0826 69.4%
- 65 nomic-ai/nomic-embed-text-v2-moe 69.4%
- 66 intfloat/e5-base-v2 69.3%
- 67
EmbeddingGemma 2 68.9%
- 68 BAAI/bge-small-en-v1.5 68.8%
- 69 NovaSearch/stella_en_400M_v5 68.8%
- 70 Lajavaness/bilingual-embedding-large 68.4%
- 71 mongodb/voyage-3-lite 68.3%
- 72 Mihaiii/gte-micro-v4 68.3%
- 73 Snowflake/snowflake-arctic-embed-l-v2.0 68.1%
- 74 Cohere/Cohere-embed-multilingual-v3.0 68%
- 75 NovaSearch/jasper_en_vision_language_v1 68%
- 76 codefuse-ai/F2LLM-v2-1.7B 67.9%
- 77 intfloat/e5-mistral-7b-instruct 67.9%
- 78 abhinand/MedEmbed-small-v0.1 67.8%
- 79 BAAI/bge-m3 67.7%
- 80 HIT-TMG/KaLM-embedding-multilingual-mini-v1 67.6%
- 81 sentence-transformers/all-MiniLM-L12-v2 67.2%
- 82 Snowflake/snowflake-arctic-embed-m-v1.5 67%
- 83 Cohere/Cohere-embed-english-v3.0 67%
- 84 thenlper/gte-small 66.8%
- 85 intfloat/multilingual-e5-large 66.8%
- 86
Qwen3-Embedding-4B 66.7%
- 87 mongodb/voyage-3 66.5%
- 88 Snowflake/snowflake-arctic-embed-m 66.5%
- 89 hotchpotch/bekko-embedding-v1-a25m 66.3%
- 90 Snowflake/snowflake-arctic-embed-l 66.2%
- 91 jinaai/jina-embedding-s-en-v1 66.2%
- 92 hotchpotch/bekko-embedding-v1-a8m 66.2%
- 93
Qwen3-Embedding-8B 66.1%
- 94 intfloat/multilingual-e5-small 65.9%
- 95 manu/sentence_croissant_alpha_v0.4 65.7%
- 96 avsolatorio/GIST-Embedding-v0 65.5%
- 97 manu/bge-m3-custom-fr 65.5%
- 98 izhx/udever-bloom-560m 65.4%
- 99 albertobarnabo/bge-m3-italian 65.3%
- 100 zeroentropy/zembed-1 65.3%
- 101 intfloat/e5-small-v2 65.1%
- 102 deepvk/USER-bge-m3 64.9%
- 103 Lajavaness/bilingual-embedding-base 64.8%
- 104 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet 64.7%
- 105 codefuse-ai/F2LLM-v2-330M 64.5%
- 106 dwzhu/e5-base-4k 64.3%
- 107 Snowflake/snowflake-arctic-embed-s 64.2%
- 108 Salesforce/SFR-Embedding-2_R 64.1%
- 109 intfloat/e5-large-v2 64%
- 110 intfloat/e5-base 63.6%
- 111 ibm-granite/granite-embedding-107m-multilingual 63.4%
- 112 BidirLM/BidirLM-270M-Embedding 63%
- 113 openai/text-embedding-3-small 63%
- 114 intfloat/multilingual-e5-large-instruct 62.9%
- 115 nomic-ai/nomic-embed-text-v1-ablated 62.9%
- 116 Alibaba-NLP/gte-multilingual-base 62.9%
- 117 nomic-ai/nomic-embed-text-v1 62.9%
- 118 jinaai/jina-embeddings-v3 62.8%
- 119 Mihaiii/Bulbasaur 62.4%
- 120 nomic-ai/nomic-embed-text-v1-unsupervised 62.4%
- 121 deepfile/embedder-100p 62.3%
- 122 codefuse-ai/F2LLM-v2-0.6B 62.2%
- 123 mongodb/voyage-3.5 62.1%
- 124 codefuse-ai/F2LLM-v2-160M 62%
- 125 manu/sentence_croissant_alpha_v0.3 62%
- 126
Qwen3-Embedding-0.6B 61.7%
- 127 intfloat/multilingual-e5-base 61.5%
- 128 Snowflake/snowflake-arctic-embed-xs 61.4%
- 129 avsolatorio/GIST-small-Embedding-v0 61.3%
- 130 google/text-multilingual-embedding-002 61.2%
- 131 avsolatorio/GIST-all-MiniLM-L6-v2 60.7%
- 132 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 60.5%
- 133 Cohere/Cohere-embed-english-light-v3.0 60.4%
- 134 sentence-transformers/all-mpnet-base-v2 59.2%
- 135 hotchpotch/bekko-static-embedding-v1 58.9%
- 136 ibm-granite/granite-embedding-30m-english 58.6%
- 137 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet 58.4%
- 138 codefuse-ai/F2LLM-v2-80M 58%
- 139 Mihaiii/Venusaur 56.6%
- 140 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka 56.4%
- 141 ibm-granite/granite-embedding-97m-multilingual-r2 55.7%
- 142 ibm-granite/granite-embedding-311m-multilingual-r2 55.4%
- 143 nomic-ai/nomic-embed-text-v1.5 55.4%
- 144 Gameselo/STS-multilingual-mpnet-base-v2 55.3%
- 145 Mihaiii/Wartortle 55.3%
- 146 sentence-transformers/static-similarity-mrl-multilingual-v1 54.6%
- 147 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 54.5%
- 148 intfloat/e5-large 54.3%
- 149 jinaai/jina-embedding-b-en-v1 53.7%
- 150 nvidia/NV-Embed-v1 53.4%
- 151 sentence-transformers/LaBSE 53.3%
- 152 omarelshehy/arabic-english-sts-matryoshka 53.2%
- 153 Jaume/gemma-2b-embeddings 53.1%
- 154 sergeyzh/LaBSE-ru-turbo 52.4%
- 155 minishlab/potion-multilingual-128M 51.9%
- 156 brahmairesearch/slx-v0.1 51.2%
- 157 sentence-transformers/all-MiniLM-L6-v2 51.2%
- 158 shibing624/text2vec-base-multilingual 50.9%
- 159 cnmoro/static-nomic-384-pten-v2-st 50.7%
- 160 intfloat/e5-small 50.4%
- 161 deepvk/USER-base 50.1%
- 162 ibm-granite/granite-embedding-125m-english 49.4%
- 163 Mihaiii/Squirtle 48.4%
- 164 consciousAI/cai-lunaris-text-embeddings 48%
- 165 Mihaiii/gte-micro 45.9%
- 166 ai-forever/ru-en-RoSBERTa 45.4%
- 167 aari1995/German_Semantic_STS_V2 44.6%
- 168 nvidia/NV-Embed-v2 43.7%
- 169 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka 42.3%
- 170 Snowflake/snowflake-arctic-embed-m-long 41.6%
- 171 cointegrated/LaBSE-en-ru 40.9%
- 172 deepvk/deberta-v1-base 40.7%
- 173 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka 40.7%
- 174 minishlab/M2V_base_glove_subword 40%
- 175 DeepPavlov/distilrubert-small-cased-conversational 36.3%
- 176 minishlab/M2V_base_output 34.9%
- 177 DeepPavlov/rubert-base-cased 34.7%
- 178 silma-ai/silma-embeddding-matryoshka-v0.1 33.2%
- 179 minishlab/M2V_base_glove 32.3%
- 180 minishlab/potion-base-8M 31.6%
- 181 DeepPavlov/rubert-base-cased-sentence 30.5%
- 182 cointegrated/rubert-tiny 30.4%
- 183 minishlab/potion-base-4M 27.4%
- 184 malenia1/ternary-weight-embedding 23.3%
- 185 ai-forever/sbert_large_nlu_ru 23.1%
- 186 minishlab/potion-base-2M 22.9%
- 187 ai-forever/sbert_large_mt_nlu_ru 22.9%
- 188 cointegrated/rubert-tiny2 18.2%
- 189 sergeyzh/rubert-tiny-turbo 18.1%
- 190 jinaai/jina-embeddings-v2-small-en 16.6%
- 191 jinaai/jina-embeddings-v2-base-en 7.33%
- 192 mteb/baseline-random-encoder 2.35%
Source: Official Italian-Administrative-CaseHold leaderboard
- 1
EmbeddingGemma 2 68.9%
- 2
Qwen3-Embedding-4B 66.7%
- 3
Qwen3-Embedding-8B 66.1%
- 4
Qwen3-Embedding-0.6B 61.7%
Source: Official Italian-Administrative-CaseHold leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | bigscience/sgpt-bloom-7b1-msmarco | 94.4% | Official leaderboard | ||
| 2 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised | 92.1% | Official leaderboard | ||
| 3 | WhereIsAI/UAE-Large-V1 | 91.7% | Official leaderboard | ||
| 4 | jinaai/jina-embeddings-v5-omni-small | 91% | Official leaderboard | ||
| 5 | jinaai/jina-embeddings-v5-text-small | 91% | Official leaderboard | ||
| 6 | mixedbread-ai/mxbai-embed-large-v1 | 89.8% | Official leaderboard | ||
| 7 | izhx/udever-bloom-7b1 | 88.5% | Official leaderboard | ||
| 8 | jinaai/jina-embeddings-v5-omni-nano | 87.4% | Official leaderboard | ||
| 9 | jinaai/jina-embeddings-v5-text-nano | 87.4% | Official leaderboard | ||
| 10 | BidirLM/BidirLM-Omni-2.5B-Embedding | 85.5% | Official leaderboard | ||
| 11 | Bytedance/Seed1.6-embedding-1215 | 85.3% | Official leaderboard | ||
| 12 | sdadas/mmlw-e5-large | 85% | Official leaderboard | ||
| 13 | tencent/KaLM-Embedding-Gemma3-12B-2511 | 85% | Official leaderboard | ||
| 14 | sdadas/mmlw-e5-base | 85% | Official leaderboard | ||
| 15 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse | 84.4% | Official leaderboard | ||
| 16 | sdadas/mmlw-roberta-base | 84.2% | Official leaderboard | ||
| 17 | sdadas/mmlw-roberta-large | 84% | Official leaderboard | ||
| 18 | Mihaiii/Ivysaur | 82.5% | Official leaderboard | ||
| 19 | sdadas/mmlw-e5-small | 81.6% | Official leaderboard | ||
| 20 | avsolatorio/NoInstruct-small-Embedding-v0 | 80.3% | Official leaderboard | ||
| 21 | izhx/udever-bloom-3b | 79.3% | Official leaderboard | ||
| 22 | Lajavaness/bilingual-embedding-small | 79.3% | Official leaderboard | ||
| 23 | microsoft/harrier-oss-v1-27b | 79.2% | Official leaderboard | ||
| 24 | microsoft/harrier-oss-v1-0.6b | 78.8% | Official leaderboard | ||
| 25 | Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka | 78.7% | Official leaderboard | ||
| 26 | izhx/udever-bloom-1b1 | 78% | Official leaderboard | ||
| 27 | BidirLM/BidirLM-1.7B-Embedding | 78% | Official leaderboard | ||
| 28 | microsoft/harrier-oss-v1-270m | 76.6% | Official leaderboard | ||
| 29 | BidirLM/BidirLM-0.6B-Embedding | 76.2% | Official leaderboard | ||
| 30 | Haon-Chen/speed-embedding-7b-instruct | 76% | Official leaderboard | ||
| 31 | BAAI/bge-large-en-v1.5 | 75.9% | Official leaderboard | ||
| 32 | GritLM/GritLM-8x7B | 75.1% | Official leaderboard | ||
| 33 | avsolatorio/GIST-large-Embedding-v0 | 74.6% | Official leaderboard | ||
| 34 | GritLM/GritLM-7B | 73.5% | Official leaderboard | ||
| 35 | google/gemini-embedding-001 | 73.3% | Official leaderboard | ||
| 36 | nvidia/llama-embed-nemotron-8b | 73.2% | Official leaderboard | ||
| 37 | codefuse-ai/F2LLM-v2-14B | 73% | Official leaderboard | ||
| 38 | Cohere/Cohere-embed-multilingual-light-v3.0 | 72.8% | Official leaderboard | ||
| 39 | thenlper/gte-large | 72.5% | Official leaderboard | ||
| 40 | Mira190/Euler-Legal-Embedding-V1 | 72.4% | Official leaderboard | ||
| 41 | Alibaba-NLP/gte-Qwen1.5-7B-instruct | 72.4% | Official leaderboard | ||
| 42 | codefuse-ai/F2LLM-v2-8B | 72.2% | Official leaderboard | ||
| 43 | ICT-TIME-and-Querit/BOOM_4B_v1 | 72.2% | Official leaderboard | ||
| 44 | ibm-granite/granite-embedding-278m-multilingual | 72% | Official leaderboard | ||
| 45 | ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 | 72% | Official leaderboard | ||
| 46 | sentence-transformers/paraphrase-multilingual-mpnet-base-v2 | 71.8% | Official leaderboard | ||
| 47 | Octen/Octen-Embedding-8B | 71.6% | Official leaderboard | ||
| 48 | Alibaba-NLP/gte-Qwen2-7B-instruct | 71.6% | Official leaderboard | ||
| 49 | NovaSearch/stella_en_1.5B_v5 | 71.6% | Official leaderboard | ||
| 50 | codefuse-ai/F2LLM-v2-4B | 71.5% | Official leaderboard | ||
| 51 | Snowflake/snowflake-arctic-embed-m-v2.0 | 71.4% | Official leaderboard | ||
| 52 | Salesforce/SFR-Embedding-Mistral | 71.3% | Official leaderboard | ||
| 53 | Linq-AI-Research/Linq-Embed-Mistral | 71% | Official leaderboard | ||
| 54 | BidirLM/BidirLM-1B-Embedding | 70.9% | Official leaderboard | ||
| 55 | infgrad/stella-base-en-v2 | 70.7% | Official leaderboard | ||
| 56 | ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 | 70.5% | Official leaderboard | ||
| 57 | google/embeddinggemma-300m | 70.4% | Official leaderboard | ||
| 58 | openai/text-embedding-3-large | 70.1% | Official leaderboard | ||
| 59 | Alibaba-NLP/gte-Qwen2-1.5B-instruct | 70.1% | Official leaderboard | ||
| 60 | manu/sentence_croissant_alpha_v0.2 | 69.9% | Official leaderboard | ||
| 61 | BAAI/bge-base-en-v1.5 | 69.7% | Official leaderboard | ||
| 62 | thenlper/gte-base | 69.7% | Official leaderboard | ||
| 63 | OrdalieTech/Solon-embeddings-large-0.1 | 69.6% | Official leaderboard | ||
| 64 | ai-sage/Giga-Embeddings-instruct-3B-0826 | 69.4% | Official leaderboard | ||
| 65 | nomic-ai/nomic-embed-text-v2-moe | 69.4% | Official leaderboard | ||
| 66 | intfloat/e5-base-v2 | 69.3% | Official leaderboard | ||
| 67 | | 68.9% | Oct 2026 | Official leaderboard | |
| 68 | BAAI/bge-small-en-v1.5 | 68.8% | Official leaderboard | ||
| 69 | NovaSearch/stella_en_400M_v5 | 68.8% | Official leaderboard | ||
| 70 | Lajavaness/bilingual-embedding-large | 68.4% | Official leaderboard | ||
| 71 | mongodb/voyage-3-lite | 68.3% | Official leaderboard | ||
| 72 | Mihaiii/gte-micro-v4 | 68.3% | Official leaderboard | ||
| 73 | Snowflake/snowflake-arctic-embed-l-v2.0 | 68.1% | Official leaderboard | ||
| 74 | Cohere/Cohere-embed-multilingual-v3.0 | 68% | Official leaderboard | ||
| 75 | NovaSearch/jasper_en_vision_language_v1 | 68% | Official leaderboard | ||
| 76 | codefuse-ai/F2LLM-v2-1.7B | 67.9% | Official leaderboard | ||
| 77 | intfloat/e5-mistral-7b-instruct | 67.9% | Official leaderboard | ||
| 78 | abhinand/MedEmbed-small-v0.1 | 67.8% | Official leaderboard | ||
| 79 | BAAI/bge-m3 | 67.7% | Official leaderboard | ||
| 80 | HIT-TMG/KaLM-embedding-multilingual-mini-v1 | 67.6% | Official leaderboard | ||
| 81 | sentence-transformers/all-MiniLM-L12-v2 | 67.2% | Official leaderboard | ||
| 82 | Snowflake/snowflake-arctic-embed-m-v1.5 | 67% | Official leaderboard | ||
| 83 | Cohere/Cohere-embed-english-v3.0 | 67% | Official leaderboard | ||
| 84 | thenlper/gte-small | 66.8% | Official leaderboard | ||
| 85 | intfloat/multilingual-e5-large | 66.8% | Official leaderboard | ||
| 86 | | 66.7% | Jun 2025 | Official leaderboard | |
| 87 | mongodb/voyage-3 | 66.5% | Official leaderboard | ||
| 88 | Snowflake/snowflake-arctic-embed-m | 66.5% | Official leaderboard | ||
| 89 | hotchpotch/bekko-embedding-v1-a25m | 66.3% | Official leaderboard | ||
| 90 | Snowflake/snowflake-arctic-embed-l | 66.2% | Official leaderboard | ||
| 91 | jinaai/jina-embedding-s-en-v1 | 66.2% | Official leaderboard | ||
| 92 | hotchpotch/bekko-embedding-v1-a8m | 66.2% | Official leaderboard | ||
| 93 | | 66.1% | Jun 2025 | Official leaderboard | |
| 94 | intfloat/multilingual-e5-small | 65.9% | Official leaderboard | ||
| 95 | manu/sentence_croissant_alpha_v0.4 | 65.7% | Official leaderboard | ||
| 96 | avsolatorio/GIST-Embedding-v0 | 65.5% | Official leaderboard | ||
| 97 | manu/bge-m3-custom-fr | 65.5% | Official leaderboard | ||
| 98 | izhx/udever-bloom-560m | 65.4% | Official leaderboard | ||
| 99 | albertobarnabo/bge-m3-italian | 65.3% | Official leaderboard | ||
| 100 | zeroentropy/zembed-1 | 65.3% | Official leaderboard | ||
| 101 | intfloat/e5-small-v2 | 65.1% | Official leaderboard | ||
| 102 | deepvk/USER-bge-m3 | 64.9% | Official leaderboard | ||
| 103 | Lajavaness/bilingual-embedding-base | 64.8% | Official leaderboard | ||
| 104 | Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet | 64.7% | Official leaderboard | ||
| 105 | codefuse-ai/F2LLM-v2-330M | 64.5% | Official leaderboard | ||
| 106 | dwzhu/e5-base-4k | 64.3% | Official leaderboard | ||
| 107 | Snowflake/snowflake-arctic-embed-s | 64.2% | Official leaderboard | ||
| 108 | Salesforce/SFR-Embedding-2_R | 64.1% | Official leaderboard | ||
| 109 | intfloat/e5-large-v2 | 64% | Official leaderboard | ||
| 110 | intfloat/e5-base | 63.6% | Official leaderboard | ||
| 111 | ibm-granite/granite-embedding-107m-multilingual | 63.4% | Official leaderboard | ||
| 112 | BidirLM/BidirLM-270M-Embedding | 63% | Official leaderboard | ||
| 113 | openai/text-embedding-3-small | 63% | Official leaderboard | ||
| 114 | intfloat/multilingual-e5-large-instruct | 62.9% | Official leaderboard | ||
| 115 | nomic-ai/nomic-embed-text-v1-ablated | 62.9% | Official leaderboard | ||
| 116 | Alibaba-NLP/gte-multilingual-base | 62.9% | Official leaderboard | ||
| 117 | nomic-ai/nomic-embed-text-v1 | 62.9% | Official leaderboard | ||
| 118 | jinaai/jina-embeddings-v3 | 62.8% | Official leaderboard | ||
| 119 | Mihaiii/Bulbasaur | 62.4% | Official leaderboard | ||
| 120 | nomic-ai/nomic-embed-text-v1-unsupervised | 62.4% | Official leaderboard | ||
| 121 | deepfile/embedder-100p | 62.3% | Official leaderboard | ||
| 122 | codefuse-ai/F2LLM-v2-0.6B | 62.2% | Official leaderboard | ||
| 123 | mongodb/voyage-3.5 | 62.1% | Official leaderboard | ||
| 124 | codefuse-ai/F2LLM-v2-160M | 62% | Official leaderboard | ||
| 125 | manu/sentence_croissant_alpha_v0.3 | 62% | Official leaderboard | ||
| 126 | | 61.7% | Jun 2025 | Official leaderboard | |
| 127 | intfloat/multilingual-e5-base | 61.5% | Official leaderboard | ||
| 128 | Snowflake/snowflake-arctic-embed-xs | 61.4% | Official leaderboard | ||
| 129 | avsolatorio/GIST-small-Embedding-v0 | 61.3% | Official leaderboard | ||
| 130 | google/text-multilingual-embedding-002 | 61.2% | Official leaderboard | ||
| 131 | avsolatorio/GIST-all-MiniLM-L6-v2 | 60.7% | Official leaderboard | ||
| 132 | HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 | 60.5% | Official leaderboard | ||
| 133 | Cohere/Cohere-embed-english-light-v3.0 | 60.4% | Official leaderboard | ||
| 134 | sentence-transformers/all-mpnet-base-v2 | 59.2% | Official leaderboard | ||
| 135 | hotchpotch/bekko-static-embedding-v1 | 58.9% | Official leaderboard | ||
| 136 | ibm-granite/granite-embedding-30m-english | 58.6% | Official leaderboard | ||
| 137 | Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet | 58.4% | Official leaderboard | ||
| 138 | codefuse-ai/F2LLM-v2-80M | 58% | Official leaderboard | ||
| 139 | Mihaiii/Venusaur | 56.6% | Official leaderboard | ||
| 140 | Omartificial-Intelligence-Space/Arabic-labse-Matryoshka | 56.4% | Official leaderboard | ||
| 141 | ibm-granite/granite-embedding-97m-multilingual-r2 | 55.7% | Official leaderboard | ||
| 142 | ibm-granite/granite-embedding-311m-multilingual-r2 | 55.4% | Official leaderboard | ||
| 143 | nomic-ai/nomic-embed-text-v1.5 | 55.4% | Official leaderboard | ||
| 144 | Gameselo/STS-multilingual-mpnet-base-v2 | 55.3% | Official leaderboard | ||
| 145 | Mihaiii/Wartortle | 55.3% | Official leaderboard | ||
| 146 | sentence-transformers/static-similarity-mrl-multilingual-v1 | 54.6% | Official leaderboard | ||
| 147 | sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 | 54.5% | Official leaderboard | ||
| 148 | intfloat/e5-large | 54.3% | Official leaderboard | ||
| 149 | jinaai/jina-embedding-b-en-v1 | 53.7% | Official leaderboard | ||
| 150 | nvidia/NV-Embed-v1 | 53.4% | Official leaderboard | ||
| 151 | sentence-transformers/LaBSE | 53.3% | Official leaderboard | ||
| 152 | omarelshehy/arabic-english-sts-matryoshka | 53.2% | Official leaderboard | ||
| 153 | Jaume/gemma-2b-embeddings | 53.1% | Official leaderboard | ||
| 154 | sergeyzh/LaBSE-ru-turbo | 52.4% | Official leaderboard | ||
| 155 | minishlab/potion-multilingual-128M | 51.9% | Official leaderboard | ||
| 156 | brahmairesearch/slx-v0.1 | 51.2% | Official leaderboard | ||
| 157 | sentence-transformers/all-MiniLM-L6-v2 | 51.2% | Official leaderboard | ||
| 158 | shibing624/text2vec-base-multilingual | 50.9% | Official leaderboard | ||
| 159 | cnmoro/static-nomic-384-pten-v2-st | 50.7% | Official leaderboard | ||
| 160 | intfloat/e5-small | 50.4% | Official leaderboard | ||
| 161 | deepvk/USER-base | 50.1% | Official leaderboard | ||
| 162 | ibm-granite/granite-embedding-125m-english | 49.4% | Official leaderboard | ||
| 163 | Mihaiii/Squirtle | 48.4% | Official leaderboard | ||
| 164 | consciousAI/cai-lunaris-text-embeddings | 48% | Official leaderboard | ||
| 165 | Mihaiii/gte-micro | 45.9% | Official leaderboard | ||
| 166 | ai-forever/ru-en-RoSBERTa | 45.4% | Official leaderboard | ||
| 167 | aari1995/German_Semantic_STS_V2 | 44.6% | Official leaderboard | ||
| 168 | nvidia/NV-Embed-v2 | 43.7% | Official leaderboard | ||
| 169 | Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka | 42.3% | Official leaderboard | ||
| 170 | Snowflake/snowflake-arctic-embed-m-long | 41.6% | Official leaderboard | ||
| 171 | cointegrated/LaBSE-en-ru | 40.9% | Official leaderboard | ||
| 172 | deepvk/deberta-v1-base | 40.7% | Official leaderboard | ||
| 173 | Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka | 40.7% | Official leaderboard | ||
| 174 | minishlab/M2V_base_glove_subword | 40% | Official leaderboard | ||
| 175 | DeepPavlov/distilrubert-small-cased-conversational | 36.3% | Official leaderboard | ||
| 176 | minishlab/M2V_base_output | 34.9% | Official leaderboard | ||
| 177 | DeepPavlov/rubert-base-cased | 34.7% | Official leaderboard | ||
| 178 | silma-ai/silma-embeddding-matryoshka-v0.1 | 33.2% | Official leaderboard | ||
| 179 | minishlab/M2V_base_glove | 32.3% | Official leaderboard | ||
| 180 | minishlab/potion-base-8M | 31.6% | Official leaderboard | ||
| 181 | DeepPavlov/rubert-base-cased-sentence | 30.5% | Official leaderboard | ||
| 182 | cointegrated/rubert-tiny | 30.4% | Official leaderboard | ||
| 183 | minishlab/potion-base-4M | 27.4% | Official leaderboard | ||
| 184 | malenia1/ternary-weight-embedding | 23.3% | Official leaderboard | ||
| 185 | ai-forever/sbert_large_nlu_ru | 23.1% | Official leaderboard | ||
| 186 | minishlab/potion-base-2M | 22.9% | Official leaderboard | ||
| 187 | ai-forever/sbert_large_mt_nlu_ru | 22.9% | Official leaderboard | ||
| 188 | cointegrated/rubert-tiny2 | 18.2% | Official leaderboard | ||
| 189 | sergeyzh/rubert-tiny-turbo | 18.1% | Official leaderboard | ||
| 190 | jinaai/jina-embeddings-v2-small-en | 16.6% | Official leaderboard | ||
| 191 | jinaai/jina-embeddings-v2-base-en | 7.33% | Official leaderboard | ||
| 192 | mteb/baseline-random-encoder | 2.35% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 68.9% | Oct 2026 | Official leaderboard | |
| 2 | | 66.7% | Jun 2025 | Official leaderboard | |
| 3 | | 66.1% | Jun 2025 | Official leaderboard | |
| 4 | | 61.7% | Jun 2025 | Official leaderboard |
2 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Italian-Administrative-CaseHold leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Italian-Administrative-CaseHold 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 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (192 entries).
- Pass Models scored 4 models we track, plus 188 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.
- Pass Headroom The best is 68.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on mteb-leaderboard-backend.hf.space: 4 rows for models we track and 188 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.