Estonian Valence Corpus
Models predict the emotional valence of Estonian news text using human annotations.
MTEB scores the task using accuracy.
- Made by
- Hille Pajupuu and colleagues
- Kind
- Reading in many languages
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Prediction
- Tasks
- All public
- Measures
- Capability
- Language
- Estonian
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Last model added
- October 2026
- Scores refreshed
- November 20th, 2026
microsoft/harrier-oss-v1-27b tops the leaderboard with 67.6% Qwen3-Embedding-8B leads with 59.9% among models released by October 2026
Best score per entry from MTEB’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 59.9%. 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
- 67.6 microsoft/harrier-oss-v1-27b
- 64.6 tencent/KaLM-Embedding-Gemma3-12B-2511
- 62.9 nvidia/llama-embed-nemotron-8b
- 59.9
Qwen3-Embedding-8B
- 59.6 codefuse-ai/F2LLM-v2-14B
- 59.6
Qwen3-Embedding-4B
- 58.7 Bytedance/Seed1.6-embedding-1215
- 58.5 codefuse-ai/F2LLM-v2-8B
- 58.5 microsoft/harrier-oss-v1-0.6b
- 57.7 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826
- 57.2 microsoft/harrier-oss-v1-270m
- 57.0 codefuse-ai/F2LLM-v2-4B
- 56.0 jinaai/jina-embeddings-v5-omni-small
- 56.0 jinaai/jina-embeddings-v5-text-small
- 55.5 ai-sage/Giga-Embeddings-instruct-3B-0826
- 53.5 google/gemini-embedding-001
- 51.9 BidirLM/BidirLM-1.7B-Embedding
- 51.2 codefuse-ai/F2LLM-v2-1.7B
- 51.0 jinaai/jina-embeddings-v5-omni-nano
- 51.0 jinaai/jina-embeddings-v5-text-nano
- 51.0 BidirLM/BidirLM-1B-Embedding
- 50.6 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1
- 49.2 ICT-TIME-and-Querit/BOOM_4B_v1
- 49.2 BidirLM/BidirLM-Omni-2.5B-Embedding
- 49.1 intfloat/multilingual-e5-large-instruct
- 48.8 jinaai/jina-embeddings-v3
- 46.2 google/text-multilingual-embedding-002
- 45.8
Gemini Embedding 2
- 45.0 omarelshehy/arabic-english-sts-matryoshka
- 44.2 Cohere/Cohere-embed-multilingual-v3.0
- 43.9 Linq-AI-Research/Linq-Embed-Mistral
- 43.7 BAAI/bge-m3
- 43.7 openai/text-embedding-3-large
- 43.4 Octen/Octen-Embedding-8B
- 43.2 mongodb/voyage-3.5
- 42.9 intfloat/multilingual-e5-large
- 42.9 codefuse-ai/F2LLM-v2-0.6B
- 42.8 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka
- 42.6 Alibaba-NLP/gte-Qwen1.5-7B-instruct
- 42.6 Alibaba-NLP/gte-Qwen2-7B-instruct
- 42.3 intfloat/e5-mistral-7b-instruct
- 42.2 BidirLM/BidirLM-0.6B-Embedding
- 42.1 deepvk/USER-bge-m3
- 42.0 zeroentropy/zembed-1
- 42.0 Lajavaness/bilingual-embedding-large
- 41.7 mongodb/voyage-3-lite
- 41.3 nomic-ai/nomic-embed-text-v2-moe
- 41.2 intfloat/multilingual-e5-base
- 41.1
Qwen3-Embedding-0.6B
- 41.1 manu/bge-m3-custom-fr
- 40.8 sentence-transformers/paraphrase-multilingual-mpnet-base-v2
- 40.7 Snowflake/snowflake-arctic-embed-l-v2.0
- 40.5 mongodb/voyage-multilingual-2
- 40.4 Mira190/Euler-Legal-Embedding-V1
- 40.4 GritLM/GritLM-8x7B
- 40.4 mongodb/voyage-3
- 40.4 OrdalieTech/Solon-embeddings-large-0.1
- 39.9 Salesforce/SFR-Embedding-Mistral
- 39.7 intfloat/multilingual-e5-small
- 39.5 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka
- 39.3 mongodb/voyage-large-2-instruct
- 39.1 Alibaba-NLP/gte-multilingual-base
- 39.0 Alibaba-NLP/gte-Qwen2-1.5B-instruct
- 38.8 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
- 38.7 openai/text-embedding-3-small
- 38.7
EmbeddingGemma 2
- 38.7 GritLM/GritLM-7B
- 38.6 Lajavaness/bilingual-embedding-base
- 38.5 Cohere/Cohere-embed-multilingual-light-v3.0
- 38.4 hotchpotch/bekko-embedding-v1-a25m
- 38.3 google/embeddinggemma-300m
- 38.3 sentence-transformers/LaBSE
- 38.0 Salesforce/SFR-Embedding-2_R
- 37.8 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet
- 36.9 hotchpotch/bekko-embedding-v1-a8m
- 36.8 ibm-granite/granite-embedding-311m-multilingual-r2
- 36.8 Gameselo/STS-multilingual-mpnet-base-v2
- 36.7 Lajavaness/bilingual-embedding-small
- 36.3 shibing624/text2vec-base-multilingual
- 36.1 BidirLM/BidirLM-270M-Embedding
- 35.9 Jaume/gemma-2b-embeddings
- 35.8 codefuse-ai/F2LLM-v2-160M
- 35.4 codefuse-ai/F2LLM-v2-330M
- 34.9 ibm-granite/granite-embedding-278m-multilingual
- 34.7 NovaSearch/jasper_en_vision_language_v1
- 34.6 HIT-TMG/KaLM-embedding-multilingual-mini-v1
- 34.6 nvidia/NV-Embed-v1
- 34.3 Haon-Chen/speed-embedding-7b-instruct
- 34.1 Snowflake/snowflake-arctic-embed-m-v2.0
- 34.0 hotchpotch/bekko-static-embedding-v1
- 33.4 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1
- 33.4 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised
- 33.3 NovaSearch/stella_en_1.5B_v5
- 33.2 minishlab/potion-multilingual-128M
- 33.2 codefuse-ai/F2LLM-v2-80M
- 33.1 ibm-granite/granite-embedding-97m-multilingual-r2
- 32.9 cointegrated/LaBSE-en-ru
- 32.6 nvidia/NV-Embed-v2
- 32.5 ibm-granite/granite-embedding-107m-multilingual
- 31.7 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse
- 31.6 sergeyzh/LaBSE-ru-turbo
- 31.6 cnmoro/static-nomic-384-pten-v2-st
- 31.6 deepfile/embedder-100p
- 31.3 ai-forever/ru-en-RoSBERTa
- 31.3 Cohere/Cohere-embed-english-v3.0
- 30.8 sdadas/mmlw-e5-large
- 30.6 dwzhu/e5-base-4k
- 30.3 Snowflake/snowflake-arctic-embed-m
- 30.3 deepvk/USER-base
- 30.2 cointegrated/rubert-tiny
- 30.2 nomic-ai/nomic-embed-text-v1-ablated
- 30.1 intfloat/e5-base
- 30.1 thenlper/gte-base
- 30.0 Snowflake/snowflake-arctic-embed-m-long
- 30.0 DeepPavlov/rubert-base-cased
- 30.0 sdadas/mmlw-e5-small
- 30.0 NovaSearch/stella_en_400M_v5
- 29.9 aari1995/German_Semantic_STS_V2
- 29.9 sdadas/mmlw-e5-base
- 29.9 nomic-ai/nomic-embed-text-v1
- 29.9 intfloat/e5-large
- 29.9 intfloat/e5-large-v2
- 29.9 Mihaiii/gte-micro-v4
- 29.7 nomic-ai/nomic-embed-text-v1-unsupervised
- 29.7 intfloat/e5-base-v2
- 29.7 thenlper/gte-large
- 29.6 deepvk/deberta-v1-base
- 29.6 silma-ai/silma-embeddding-matryoshka-v0.1
- 29.5 Mihaiii/Bulbasaur
- 29.5 Snowflake/snowflake-arctic-embed-l
- 29.5 bigscience/sgpt-bloom-7b1-msmarco
- 29.4 infgrad/stella-base-en-v2
- 29.4 DeepPavlov/rubert-base-cased-sentence
- 29.4 manu/sentence_croissant_alpha_v0.4
- 29.4 avsolatorio/GIST-Embedding-v0
- 29.4 avsolatorio/GIST-small-Embedding-v0
- 29.3 BAAI/bge-large-en-v1.5
- 29.3 sdadas/mmlw-roberta-large
- 29.3 Cohere/Cohere-embed-english-light-v3.0
- 29.3 intfloat/e5-small
- 29.3 DeepPavlov/distilrubert-small-cased-conversational
- 29.2 nomic-ai/nomic-embed-text-v1.5
- 29.2 avsolatorio/NoInstruct-small-Embedding-v0
- 29.1 mixedbread-ai/mxbai-embed-large-v1
- 29.1 sdadas/mmlw-roberta-base
- 29.1 izhx/udever-bloom-7b1
- 29.1 avsolatorio/GIST-large-Embedding-v0
- 29.0 cointegrated/rubert-tiny2
- 29.0 Mihaiii/Venusaur
- 29.0 Mihaiii/gte-micro
- 29.0 WhereIsAI/UAE-Large-V1
- 28.9 BAAI/bge-base-en-v1.5
- 28.9 Snowflake/snowflake-arctic-embed-s
- 28.9 jinaai/jina-embeddings-v2-small-en
- 28.9 BAAI/bge-small-en-v1.5
- 28.9 brahmairesearch/slx-v0.1
- 28.9 sentence-transformers/all-MiniLM-L6-v2
- 28.8 sentence-transformers/all-mpnet-base-v2
- 28.8 izhx/udever-bloom-3b
- 28.8 thenlper/gte-small
- 28.8 Snowflake/snowflake-arctic-embed-m-v1.5
- 28.8 Mihaiii/Wartortle
- 28.7 manu/sentence_croissant_alpha_v0.2
- 28.7 Mihaiii/Ivysaur
- 28.6 izhx/udever-bloom-1b1
- 28.6 Mihaiii/Squirtle
- 28.5 izhx/udever-bloom-560m
- 28.5 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka
- 28.4 Snowflake/snowflake-arctic-embed-xs
- 28.3 avsolatorio/GIST-all-MiniLM-L6-v2
- 28.3 ibm-granite/granite-embedding-30m-english
- 28.3 intfloat/e5-small-v2
- 28.3 abhinand/MedEmbed-small-v0.1
- 28.3 ai-forever/sbert_large_mt_nlu_ru
- 28.3 sergeyzh/rubert-tiny-turbo
- 28.3 manu/sentence_croissant_alpha_v0.3
- 28.1 sentence-transformers/all-MiniLM-L12-v2
- 28.0 ibm-granite/granite-embedding-125m-english
- 28.0 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka
- 27.6 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet
- 27.5 malenia1/ternary-weight-embedding
- 27.3 minishlab/potion-base-8M
- 27.1 minishlab/potion-base-2M
- 26.9 minishlab/M2V_base_glove
- 26.9 minishlab/potion-base-4M
- 26.9 minishlab/M2V_base_glove_subword
- 26.8 ai-forever/sbert_large_nlu_ru
- 26.8 consciousAI/cai-lunaris-text-embeddings
- 26.8 jinaai/jina-embeddings-v2-base-en
- 26.7 sentence-transformers/static-similarity-mrl-multilingual-v1
- 26.1 jinaai/jina-embedding-b-en-v1
- 25.9 minishlab/M2V_base_output
- 25.7 jinaai/jina-embedding-s-en-v1
- 1 microsoft/harrier-oss-v1-27b 67.6%
- 2 tencent/KaLM-Embedding-Gemma3-12B-2511 64.6%
- 3 nvidia/llama-embed-nemotron-8b 62.9%
- 4
Qwen3-Embedding-8B 59.9%
- 5 codefuse-ai/F2LLM-v2-14B 59.6%
- 6
Qwen3-Embedding-4B 59.6%
- 7 Bytedance/Seed1.6-embedding-1215 58.7%
- 8 codefuse-ai/F2LLM-v2-8B 58.5%
- 9 microsoft/harrier-oss-v1-0.6b 58.5%
- 10 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 57.7%
- 11 microsoft/harrier-oss-v1-270m 57.2%
- 12 codefuse-ai/F2LLM-v2-4B 57%
- 13 jinaai/jina-embeddings-v5-omni-small 56%
- 14 jinaai/jina-embeddings-v5-text-small 56%
- 15 ai-sage/Giga-Embeddings-instruct-3B-0826 55.5%
- 16 google/gemini-embedding-001 53.5%
- 17 BidirLM/BidirLM-1.7B-Embedding 51.9%
- 18 codefuse-ai/F2LLM-v2-1.7B 51.2%
- 19 jinaai/jina-embeddings-v5-omni-nano 51%
- 20 jinaai/jina-embeddings-v5-text-nano 51%
- 21 BidirLM/BidirLM-1B-Embedding 51%
- 22 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 50.6%
- 23 ICT-TIME-and-Querit/BOOM_4B_v1 49.2%
- 24 BidirLM/BidirLM-Omni-2.5B-Embedding 49.2%
- 25 intfloat/multilingual-e5-large-instruct 49.1%
- 26 jinaai/jina-embeddings-v3 48.8%
- 27 google/text-multilingual-embedding-002 46.2%
- 28
Gemini Embedding 2 45.8%
- 29 omarelshehy/arabic-english-sts-matryoshka 45%
- 30 Cohere/Cohere-embed-multilingual-v3.0 44.2%
- 31 Linq-AI-Research/Linq-Embed-Mistral 43.9%
- 32 BAAI/bge-m3 43.7%
- 33 openai/text-embedding-3-large 43.7%
- 34 Octen/Octen-Embedding-8B 43.4%
- 35 mongodb/voyage-3.5 43.2%
- 36 intfloat/multilingual-e5-large 42.9%
- 37 codefuse-ai/F2LLM-v2-0.6B 42.9%
- 38 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka 42.8%
- 39 Alibaba-NLP/gte-Qwen1.5-7B-instruct 42.6%
- 40 Alibaba-NLP/gte-Qwen2-7B-instruct 42.6%
- 41 intfloat/e5-mistral-7b-instruct 42.3%
- 42 BidirLM/BidirLM-0.6B-Embedding 42.2%
- 43 deepvk/USER-bge-m3 42.1%
- 44 zeroentropy/zembed-1 42%
- 45 Lajavaness/bilingual-embedding-large 42%
- 46 mongodb/voyage-3-lite 41.7%
- 47 nomic-ai/nomic-embed-text-v2-moe 41.3%
- 48 intfloat/multilingual-e5-base 41.2%
- 49
Qwen3-Embedding-0.6B 41.1%
- 50 manu/bge-m3-custom-fr 41.1%
- 51 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 40.8%
- 52 Snowflake/snowflake-arctic-embed-l-v2.0 40.7%
- 53 mongodb/voyage-multilingual-2 40.5%
- 54 Mira190/Euler-Legal-Embedding-V1 40.4%
- 55 GritLM/GritLM-8x7B 40.4%
- 56 mongodb/voyage-3 40.4%
- 57 OrdalieTech/Solon-embeddings-large-0.1 40.4%
- 58 Salesforce/SFR-Embedding-Mistral 39.9%
- 59 intfloat/multilingual-e5-small 39.7%
- 60 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka 39.5%
- 61 mongodb/voyage-large-2-instruct 39.3%
- 62 Alibaba-NLP/gte-multilingual-base 39.1%
- 63 Alibaba-NLP/gte-Qwen2-1.5B-instruct 39%
- 64 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 38.8%
- 65 openai/text-embedding-3-small 38.7%
- 66
EmbeddingGemma 2 38.7%
- 67 GritLM/GritLM-7B 38.7%
- 68 Lajavaness/bilingual-embedding-base 38.6%
- 69 Cohere/Cohere-embed-multilingual-light-v3.0 38.5%
- 70 hotchpotch/bekko-embedding-v1-a25m 38.4%
- 71 google/embeddinggemma-300m 38.3%
- 72 sentence-transformers/LaBSE 38.3%
- 73 Salesforce/SFR-Embedding-2_R 38%
- 74 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet 37.8%
- 75 hotchpotch/bekko-embedding-v1-a8m 36.9%
- 76 ibm-granite/granite-embedding-311m-multilingual-r2 36.8%
- 77 Gameselo/STS-multilingual-mpnet-base-v2 36.8%
- 78 Lajavaness/bilingual-embedding-small 36.7%
- 79 shibing624/text2vec-base-multilingual 36.3%
- 80 BidirLM/BidirLM-270M-Embedding 36.1%
- 81 Jaume/gemma-2b-embeddings 35.9%
- 82 codefuse-ai/F2LLM-v2-160M 35.8%
- 83 codefuse-ai/F2LLM-v2-330M 35.4%
- 84 ibm-granite/granite-embedding-278m-multilingual 34.9%
- 85 NovaSearch/jasper_en_vision_language_v1 34.7%
- 86 HIT-TMG/KaLM-embedding-multilingual-mini-v1 34.6%
- 87 nvidia/NV-Embed-v1 34.6%
- 88 Haon-Chen/speed-embedding-7b-instruct 34.3%
- 89 Snowflake/snowflake-arctic-embed-m-v2.0 34.1%
- 90 hotchpotch/bekko-static-embedding-v1 34%
- 91 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 33.4%
- 92 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised 33.4%
- 93 NovaSearch/stella_en_1.5B_v5 33.3%
- 94 minishlab/potion-multilingual-128M 33.2%
- 95 codefuse-ai/F2LLM-v2-80M 33.2%
- 96 ibm-granite/granite-embedding-97m-multilingual-r2 33.1%
- 97 cointegrated/LaBSE-en-ru 32.9%
- 98 nvidia/NV-Embed-v2 32.6%
- 99 ibm-granite/granite-embedding-107m-multilingual 32.5%
- 100 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse 31.7%
- 101 sergeyzh/LaBSE-ru-turbo 31.6%
- 102 cnmoro/static-nomic-384-pten-v2-st 31.6%
- 103 deepfile/embedder-100p 31.6%
- 104 ai-forever/ru-en-RoSBERTa 31.3%
- 105 Cohere/Cohere-embed-english-v3.0 31.3%
- 106 sdadas/mmlw-e5-large 30.8%
- 107 dwzhu/e5-base-4k 30.6%
- 108 Snowflake/snowflake-arctic-embed-m 30.3%
- 109 deepvk/USER-base 30.3%
- 110 cointegrated/rubert-tiny 30.2%
- 111 nomic-ai/nomic-embed-text-v1-ablated 30.2%
- 112 intfloat/e5-base 30.1%
- 113 thenlper/gte-base 30.1%
- 114 Snowflake/snowflake-arctic-embed-m-long 30%
- 115 DeepPavlov/rubert-base-cased 30%
- 116 sdadas/mmlw-e5-small 30%
- 117 NovaSearch/stella_en_400M_v5 30%
- 118 aari1995/German_Semantic_STS_V2 29.9%
- 119 sdadas/mmlw-e5-base 29.9%
- 120 nomic-ai/nomic-embed-text-v1 29.9%
- 121 intfloat/e5-large 29.9%
- 122 intfloat/e5-large-v2 29.9%
- 123 Mihaiii/gte-micro-v4 29.9%
- 124 nomic-ai/nomic-embed-text-v1-unsupervised 29.7%
- 125 intfloat/e5-base-v2 29.7%
- 126 thenlper/gte-large 29.7%
- 127 deepvk/deberta-v1-base 29.6%
- 128 silma-ai/silma-embeddding-matryoshka-v0.1 29.6%
- 129 Mihaiii/Bulbasaur 29.5%
- 130 Snowflake/snowflake-arctic-embed-l 29.5%
- 131 bigscience/sgpt-bloom-7b1-msmarco 29.5%
- 132 infgrad/stella-base-en-v2 29.4%
- 133 DeepPavlov/rubert-base-cased-sentence 29.4%
- 134 manu/sentence_croissant_alpha_v0.4 29.4%
- 135 avsolatorio/GIST-Embedding-v0 29.4%
- 136 avsolatorio/GIST-small-Embedding-v0 29.4%
- 137 BAAI/bge-large-en-v1.5 29.3%
- 138 sdadas/mmlw-roberta-large 29.3%
- 139 Cohere/Cohere-embed-english-light-v3.0 29.3%
- 140 intfloat/e5-small 29.3%
- 141 DeepPavlov/distilrubert-small-cased-conversational 29.3%
- 142 nomic-ai/nomic-embed-text-v1.5 29.2%
- 143 avsolatorio/NoInstruct-small-Embedding-v0 29.2%
- 144 mixedbread-ai/mxbai-embed-large-v1 29.1%
- 145 sdadas/mmlw-roberta-base 29.1%
- 146 izhx/udever-bloom-7b1 29.1%
- 147 avsolatorio/GIST-large-Embedding-v0 29.1%
- 148 cointegrated/rubert-tiny2 29%
- 149 Mihaiii/Venusaur 29%
- 150 Mihaiii/gte-micro 29%
- 151 WhereIsAI/UAE-Large-V1 29%
- 152 BAAI/bge-base-en-v1.5 28.9%
- 153 Snowflake/snowflake-arctic-embed-s 28.9%
- 154 jinaai/jina-embeddings-v2-small-en 28.9%
- 155 BAAI/bge-small-en-v1.5 28.9%
- 156 brahmairesearch/slx-v0.1 28.9%
- 157 sentence-transformers/all-MiniLM-L6-v2 28.9%
- 158 sentence-transformers/all-mpnet-base-v2 28.8%
- 159 izhx/udever-bloom-3b 28.8%
- 160 thenlper/gte-small 28.8%
- 161 Snowflake/snowflake-arctic-embed-m-v1.5 28.8%
- 162 Mihaiii/Wartortle 28.8%
- 163 manu/sentence_croissant_alpha_v0.2 28.7%
- 164 Mihaiii/Ivysaur 28.7%
- 165 izhx/udever-bloom-1b1 28.6%
- 166 Mihaiii/Squirtle 28.6%
- 167 izhx/udever-bloom-560m 28.5%
- 168 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka 28.5%
- 169 Snowflake/snowflake-arctic-embed-xs 28.4%
- 170 avsolatorio/GIST-all-MiniLM-L6-v2 28.3%
- 171 ibm-granite/granite-embedding-30m-english 28.3%
- 172 intfloat/e5-small-v2 28.3%
- 173 abhinand/MedEmbed-small-v0.1 28.3%
- 174 ai-forever/sbert_large_mt_nlu_ru 28.3%
- 175 sergeyzh/rubert-tiny-turbo 28.3%
- 176 manu/sentence_croissant_alpha_v0.3 28.3%
- 177 sentence-transformers/all-MiniLM-L12-v2 28.1%
- 178 ibm-granite/granite-embedding-125m-english 28%
- 179 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka 28%
- 180 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet 27.6%
- 181 malenia1/ternary-weight-embedding 27.5%
- 182 minishlab/potion-base-8M 27.3%
- 183 minishlab/potion-base-2M 27.1%
- 184 minishlab/M2V_base_glove 26.9%
- 185 minishlab/potion-base-4M 26.9%
- 186 minishlab/M2V_base_glove_subword 26.9%
- 187 ai-forever/sbert_large_nlu_ru 26.8%
- 188 consciousAI/cai-lunaris-text-embeddings 26.8%
- 189 jinaai/jina-embeddings-v2-base-en 26.8%
- 190 sentence-transformers/static-similarity-mrl-multilingual-v1 26.7%
- 191 jinaai/jina-embedding-b-en-v1 26.1%
- 192 minishlab/M2V_base_output 25.9%
- 193 jinaai/jina-embedding-s-en-v1 25.7%
Source: MTEB’s leaderboard
- 1
Qwen3-Embedding-8B 59.9%
- 2
Qwen3-Embedding-4B 59.6%
- 3
Gemini Embedding 2 45.8%
- 4
Qwen3-Embedding-0.6B 41.1%
- 5
EmbeddingGemma 2 38.7%
Source: MTEB’s leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | microsoft/harrier-oss-v1-27b open weights, 27B | 67.6% | huggingface.co | Official leaderboard | ||
| 2 | tencent/KaLM-Embedding-Gemma3-12B-2511 open weights, 11.8B | 64.6% | huggingface.co | Official leaderboard | ||
| 3 | nvidia/llama-embed-nemotron-8b open weights, 7.5B | 62.9% | huggingface.co | Official leaderboard | ||
| 4 | | 59.9% | Jun 2025 | huggingface.co | Official leaderboard | |
| 5 | codefuse-ai/F2LLM-v2-14B open weights, 14B | 59.6% | huggingface.co | Official leaderboard | ||
| 6 | | 59.6% | Jun 2025 | huggingface.co | Official leaderboard | |
| 7 | Bytedance/Seed1.6-embedding-1215 closed weights | 58.7% | console.volcengine.com | Official leaderboard | ||
| 8 | codefuse-ai/F2LLM-v2-8B open weights, 7.6B | 58.5% | huggingface.co | Official leaderboard | ||
| 9 | microsoft/harrier-oss-v1-0.6b open weights, 0.6B | 58.5% | huggingface.co | Official leaderboard | ||
| 10 | ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 open weights, 10.5B | 57.7% | huggingface.co | Official leaderboard | ||
| 11 | microsoft/harrier-oss-v1-270m open weights, 0.3B | 57.2% | huggingface.co | Official leaderboard | ||
| 12 | codefuse-ai/F2LLM-v2-4B open weights, 4B | 57% | huggingface.co | Official leaderboard | ||
| 13 | jinaai/jina-embeddings-v5-omni-small open weights, 1.6B | 56% | huggingface.co | Official leaderboard | ||
| 14 | jinaai/jina-embeddings-v5-text-small open weights, 0.6B | 56% | huggingface.co | Official leaderboard | ||
| 15 | ai-sage/Giga-Embeddings-instruct-3B-0826 open weights, 3.2B | 55.5% | huggingface.co | Official leaderboard | ||
| 16 | google/gemini-embedding-001 closed weights | 53.5% | ai.google.dev | Official leaderboard | ||
| 17 | BidirLM/BidirLM-1.7B-Embedding open weights, 1.7B | 51.9% | huggingface.co | Official leaderboard | ||
| 18 | codefuse-ai/F2LLM-v2-1.7B open weights, 1.7B | 51.2% | huggingface.co | Official leaderboard | ||
| 19 | jinaai/jina-embeddings-v5-omni-nano open weights, 1B | 51% | huggingface.co | Official leaderboard | ||
| 20 | jinaai/jina-embeddings-v5-text-nano open weights, 0.2B | 51% | huggingface.co | Official leaderboard | ||
| 21 | BidirLM/BidirLM-1B-Embedding open weights, 1B | 51% | huggingface.co | Official leaderboard | ||
| 22 | ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 open weights, 4B | 50.6% | huggingface.co | Official leaderboard | ||
| 23 | ICT-TIME-and-Querit/BOOM_4B_v1 open weights, 4B | 49.2% | huggingface.co | Official leaderboard | ||
| 24 | BidirLM/BidirLM-Omni-2.5B-Embedding open weights, 2.4B | 49.2% | huggingface.co | Official leaderboard | ||
| 25 | intfloat/multilingual-e5-large-instruct open weights, 0.6B | 49.1% | huggingface.co | Official leaderboard | ||
| 26 | jinaai/jina-embeddings-v3 open weights, 0.6B | 48.8% | huggingface.co | Official leaderboard | ||
| 27 | google/text-multilingual-embedding-002 closed weights | 46.2% | cloud.google.com | Official leaderboard | ||
| 28 | | 45.8% | Mar 2026 | ai.google.dev | Official leaderboard | |
| 29 | omarelshehy/arabic-english-sts-matryoshka open weights, 0.6B | 45% | huggingface.co | Official leaderboard | ||
| 30 | Cohere/Cohere-embed-multilingual-v3.0 closed weights | 44.2% | cohere.com | Official leaderboard | ||
| 31 | Linq-AI-Research/Linq-Embed-Mistral open weights, 7.1B | 43.9% | huggingface.co | Official leaderboard | ||
| 32 | BAAI/bge-m3 open weights, 0.6B | 43.7% | huggingface.co | Official leaderboard | ||
| 33 | openai/text-embedding-3-large closed weights | 43.7% | openai.com | Official leaderboard | ||
| 34 | Octen/Octen-Embedding-8B open weights, 7.6B | 43.4% | huggingface.co | Official leaderboard | ||
| 35 | mongodb/voyage-3.5 closed weights | 43.2% | blog.voyageai.com | Official leaderboard | ||
| 36 | intfloat/multilingual-e5-large open weights, 0.6B | 42.9% | huggingface.co | Official leaderboard | ||
| 37 | codefuse-ai/F2LLM-v2-0.6B open weights, 0.6B | 42.9% | huggingface.co | Official leaderboard | ||
| 38 | Omartificial-Intelligence-Space/Arabic-labse-Matryoshka open weights, 0.5B | 42.8% | huggingface.co | Official leaderboard | ||
| 39 | Alibaba-NLP/gte-Qwen1.5-7B-instruct open weights, 7.1B | 42.6% | huggingface.co | Official leaderboard | ||
| 40 | Alibaba-NLP/gte-Qwen2-7B-instruct open weights, 7.1B | 42.6% | huggingface.co | Official leaderboard | ||
| 41 | intfloat/e5-mistral-7b-instruct open weights, 7.1B | 42.3% | huggingface.co | Official leaderboard | ||
| 42 | BidirLM/BidirLM-0.6B-Embedding open weights, 0.6B | 42.2% | huggingface.co | Official leaderboard | ||
| 43 | deepvk/USER-bge-m3 open weights, 0.4B | 42.1% | huggingface.co | Official leaderboard | ||
| 44 | zeroentropy/zembed-1 open weights, 4B | 42% | huggingface.co | Official leaderboard | ||
| 45 | Lajavaness/bilingual-embedding-large open weights, 0.6B | 42% | huggingface.co | Official leaderboard | ||
| 46 | mongodb/voyage-3-lite closed weights | 41.7% | blog.voyageai.com | Official leaderboard | ||
| 47 | nomic-ai/nomic-embed-text-v2-moe open weights, 0.5B | 41.3% | huggingface.co | Official leaderboard | ||
| 48 | intfloat/multilingual-e5-base open weights, 0.3B | 41.2% | huggingface.co | Official leaderboard | ||
| 49 | | 41.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 50 | manu/bge-m3-custom-fr open weights, 0.6B | 41.1% | huggingface.co | Official leaderboard | ||
| 51 | sentence-transformers/paraphrase-multilingual-mpnet-base-v2 open weights, 0.3B | 40.8% | huggingface.co | Official leaderboard | ||
| 52 | Snowflake/snowflake-arctic-embed-l-v2.0 open weights, 0.6B | 40.7% | huggingface.co | Official leaderboard | ||
| 53 | mongodb/voyage-multilingual-2 closed weights | 40.5% | blog.voyageai.com | Official leaderboard | ||
| 54 | Mira190/Euler-Legal-Embedding-V1 open weights, 7.6B | 40.4% | huggingface.co | Official leaderboard | ||
| 55 | GritLM/GritLM-8x7B open weights, 57.9B | 40.4% | huggingface.co | Official leaderboard | ||
| 56 | mongodb/voyage-3 closed weights | 40.4% | blog.voyageai.com | Official leaderboard | ||
| 57 | OrdalieTech/Solon-embeddings-large-0.1 open weights, 0.6B | 40.4% | huggingface.co | Official leaderboard | ||
| 58 | Salesforce/SFR-Embedding-Mistral open weights, 7.1B | 39.9% | huggingface.co | Official leaderboard | ||
| 59 | intfloat/multilingual-e5-small open weights, 0.1B | 39.7% | huggingface.co | Official leaderboard | ||
| 60 | Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka open weights, 0.3B | 39.5% | huggingface.co | Official leaderboard | ||
| 61 | mongodb/voyage-large-2-instruct closed weights | 39.3% | blog.voyageai.com | Official leaderboard | ||
| 62 | Alibaba-NLP/gte-multilingual-base open weights, 0.3B | 39.1% | huggingface.co | Official leaderboard | ||
| 63 | Alibaba-NLP/gte-Qwen2-1.5B-instruct open weights, 1.5B | 39% | huggingface.co | Official leaderboard | ||
| 64 | sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 open weights, 0.1B | 38.8% | huggingface.co | Official leaderboard | ||
| 65 | openai/text-embedding-3-small closed weights | 38.7% | openai.com | Official leaderboard | ||
| 66 | | 38.7% | Oct 2026 | huggingface.co | Official leaderboard | |
| 67 | GritLM/GritLM-7B open weights, 7.2B | 38.7% | huggingface.co | Official leaderboard | ||
| 68 | Lajavaness/bilingual-embedding-base open weights, 0.3B | 38.6% | huggingface.co | Official leaderboard | ||
| 69 | Cohere/Cohere-embed-multilingual-light-v3.0 closed weights | 38.5% | cohere.com | Official leaderboard | ||
| 70 | hotchpotch/bekko-embedding-v1-a25m open weights, 0.1B | 38.4% | huggingface.co | Official leaderboard | ||
| 71 | google/embeddinggemma-300m open weights, 0.3B | 38.3% | ai.google.dev | Official leaderboard | ||
| 72 | sentence-transformers/LaBSE open weights, 0.5B | 38.3% | huggingface.co | Official leaderboard | ||
| 73 | Salesforce/SFR-Embedding-2_R open weights, 7.1B | 38% | huggingface.co | Official leaderboard | ||
| 74 | Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet open weights, 0.1B | 37.8% | huggingface.co | Official leaderboard | ||
| 75 | hotchpotch/bekko-embedding-v1-a8m open weights, 0.1B | 36.9% | huggingface.co | Official leaderboard | ||
| 76 | ibm-granite/granite-embedding-311m-multilingual-r2 open weights, 0.3B | 36.8% | huggingface.co | Official leaderboard | ||
| 77 | Gameselo/STS-multilingual-mpnet-base-v2 open weights, 0.3B | 36.8% | huggingface.co | Official leaderboard | ||
| 78 | Lajavaness/bilingual-embedding-small open weights, 0.1B | 36.7% | huggingface.co | Official leaderboard | ||
| 79 | shibing624/text2vec-base-multilingual open weights, 0.1B | 36.3% | huggingface.co | Official leaderboard | ||
| 80 | BidirLM/BidirLM-270M-Embedding open weights, 0.3B | 36.1% | huggingface.co | Official leaderboard | ||
| 81 | Jaume/gemma-2b-embeddings open weights, 2.5B | 35.9% | huggingface.co | Official leaderboard | ||
| 82 | codefuse-ai/F2LLM-v2-160M open weights, 0.2B | 35.8% | huggingface.co | Official leaderboard | ||
| 83 | codefuse-ai/F2LLM-v2-330M open weights, 0.3B | 35.4% | huggingface.co | Official leaderboard | ||
| 84 | ibm-granite/granite-embedding-278m-multilingual open weights, 0.3B | 34.9% | huggingface.co | Official leaderboard | ||
| 85 | NovaSearch/jasper_en_vision_language_v1 open weights, 1.6B | 34.7% | huggingface.co | Official leaderboard | ||
| 86 | HIT-TMG/KaLM-embedding-multilingual-mini-v1 open weights, 0.5B | 34.6% | huggingface.co | Official leaderboard | ||
| 87 | nvidia/NV-Embed-v1 open weights, 7.9B | 34.6% | huggingface.co | Official leaderboard | ||
| 88 | Haon-Chen/speed-embedding-7b-instruct open weights, 7.1B | 34.3% | huggingface.co | Official leaderboard | ||
| 89 | Snowflake/snowflake-arctic-embed-m-v2.0 open weights, 0.3B | 34.1% | huggingface.co | Official leaderboard | ||
| 90 | hotchpotch/bekko-static-embedding-v1 open weights, 0.2B | 34% | huggingface.co | Official leaderboard | ||
| 91 | HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 open weights, 0.5B | 33.4% | huggingface.co | Official leaderboard | ||
| 92 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised open weights, 7.2B | 33.4% | huggingface.co | Official leaderboard | ||
| 93 | NovaSearch/stella_en_1.5B_v5 open weights, 1.5B | 33.3% | huggingface.co | Official leaderboard | ||
| 94 | minishlab/potion-multilingual-128M open weights, 0.1B | 33.2% | huggingface.co | Official leaderboard | ||
| 95 | codefuse-ai/F2LLM-v2-80M open weights, 0.1B | 33.2% | huggingface.co | Official leaderboard | ||
| 96 | ibm-granite/granite-embedding-97m-multilingual-r2 open weights, 0.1B | 33.1% | huggingface.co | Official leaderboard | ||
| 97 | cointegrated/LaBSE-en-ru open weights, 0.1B | 32.9% | huggingface.co | Official leaderboard | ||
| 98 | nvidia/NV-Embed-v2 open weights, 7.9B | 32.6% | huggingface.co | Official leaderboard | ||
| 99 | ibm-granite/granite-embedding-107m-multilingual open weights, 0.1B | 32.5% | huggingface.co | Official leaderboard | ||
| 100 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse open weights, 7.2B | 31.7% | huggingface.co | Official leaderboard | ||
| 101 | sergeyzh/LaBSE-ru-turbo open weights, 0.1B | 31.6% | huggingface.co | Official leaderboard | ||
| 102 | cnmoro/static-nomic-384-pten-v2-st open weights, 0.1B | 31.6% | huggingface.co | Official leaderboard | ||
| 103 | deepfile/embedder-100p open weights, 0.3B | 31.6% | huggingface.co | Official leaderboard | ||
| 104 | ai-forever/ru-en-RoSBERTa open weights, 0.4B | 31.3% | huggingface.co | Official leaderboard | ||
| 105 | Cohere/Cohere-embed-english-v3.0 closed weights | 31.3% | cohere.com | Official leaderboard | ||
| 106 | sdadas/mmlw-e5-large open weights, 0.6B | 30.8% | huggingface.co | Official leaderboard | ||
| 107 | dwzhu/e5-base-4k open weights, 0.1B | 30.6% | huggingface.co | Official leaderboard | ||
| 108 | Snowflake/snowflake-arctic-embed-m open weights, 0.1B | 30.3% | huggingface.co | Official leaderboard | ||
| 109 | deepvk/USER-base open weights, 0.4B | 30.3% | huggingface.co | Official leaderboard | ||
| 110 | cointegrated/rubert-tiny open weights, 0B | 30.2% | huggingface.co | Official leaderboard | ||
| 111 | nomic-ai/nomic-embed-text-v1-ablated open weights, 0.1B | 30.2% | huggingface.co | Official leaderboard | ||
| 112 | intfloat/e5-base open weights, 0.1B | 30.1% | huggingface.co | Official leaderboard | ||
| 113 | thenlper/gte-base open weights, 0.1B | 30.1% | huggingface.co | Official leaderboard | ||
| 114 | Snowflake/snowflake-arctic-embed-m-long open weights, 0.1B | 30% | huggingface.co | Official leaderboard | ||
| 115 | DeepPavlov/rubert-base-cased open weights, 1.3B | 30% | huggingface.co | Official leaderboard | ||
| 116 | sdadas/mmlw-e5-small open weights, 0.1B | 30% | huggingface.co | Official leaderboard | ||
| 117 | NovaSearch/stella_en_400M_v5 open weights, 0.4B | 30% | huggingface.co | Official leaderboard | ||
| 118 | aari1995/German_Semantic_STS_V2 open weights, 0.3B | 29.9% | huggingface.co | Official leaderboard | ||
| 119 | sdadas/mmlw-e5-base open weights, 0.3B | 29.9% | huggingface.co | Official leaderboard | ||
| 120 | nomic-ai/nomic-embed-text-v1 open weights, 0.1B | 29.9% | huggingface.co | Official leaderboard | ||
| 121 | intfloat/e5-large open weights, 0.3B | 29.9% | huggingface.co | Official leaderboard | ||
| 122 | intfloat/e5-large-v2 open weights, 0.3B | 29.9% | huggingface.co | Official leaderboard | ||
| 123 | Mihaiii/gte-micro-v4 open weights, 0B | 29.9% | huggingface.co | Official leaderboard | ||
| 124 | nomic-ai/nomic-embed-text-v1-unsupervised open weights, 0.1B | 29.7% | huggingface.co | Official leaderboard | ||
| 125 | intfloat/e5-base-v2 open weights, 0.1B | 29.7% | huggingface.co | Official leaderboard | ||
| 126 | thenlper/gte-large open weights, 0.3B | 29.7% | huggingface.co | Official leaderboard | ||
| 127 | deepvk/deberta-v1-base open weights, 0.1B | 29.6% | huggingface.co | Official leaderboard | ||
| 128 | silma-ai/silma-embeddding-matryoshka-v0.1 open weights, 0.1B | 29.6% | huggingface.co | Official leaderboard | ||
| 129 | Mihaiii/Bulbasaur open weights, 0B | 29.5% | huggingface.co | Official leaderboard | ||
| 130 | Snowflake/snowflake-arctic-embed-l open weights, 0.3B | 29.5% | huggingface.co | Official leaderboard | ||
| 131 | bigscience/sgpt-bloom-7b1-msmarco open weights, 7.1B | 29.5% | huggingface.co | Official leaderboard | ||
| 132 | infgrad/stella-base-en-v2 open weights, 0.1B | 29.4% | huggingface.co | Official leaderboard | ||
| 133 | DeepPavlov/rubert-base-cased-sentence open weights, 0.1B | 29.4% | huggingface.co | Official leaderboard | ||
| 134 | manu/sentence_croissant_alpha_v0.4 open weights, 1.3B | 29.4% | huggingface.co | Official leaderboard | ||
| 135 | avsolatorio/GIST-Embedding-v0 open weights, 0.1B | 29.4% | huggingface.co | Official leaderboard | ||
| 136 | avsolatorio/GIST-small-Embedding-v0 open weights, 0B | 29.4% | huggingface.co | Official leaderboard | ||
| 137 | BAAI/bge-large-en-v1.5 open weights, 0.3B | 29.3% | huggingface.co | Official leaderboard | ||
| 138 | sdadas/mmlw-roberta-large open weights, 0.4B | 29.3% | huggingface.co | Official leaderboard | ||
| 139 | Cohere/Cohere-embed-english-light-v3.0 closed weights | 29.3% | cohere.com | Official leaderboard | ||
| 140 | intfloat/e5-small open weights, 0B | 29.3% | huggingface.co | Official leaderboard | ||
| 141 | DeepPavlov/distilrubert-small-cased-conversational open weights, 0.1B | 29.3% | huggingface.co | Official leaderboard | ||
| 142 | nomic-ai/nomic-embed-text-v1.5 open weights, 0.1B | 29.2% | huggingface.co | Official leaderboard | ||
| 143 | avsolatorio/NoInstruct-small-Embedding-v0 open weights, 0B | 29.2% | huggingface.co | Official leaderboard | ||
| 144 | mixedbread-ai/mxbai-embed-large-v1 open weights, 0.3B | 29.1% | huggingface.co | Official leaderboard | ||
| 145 | sdadas/mmlw-roberta-base open weights, 0.1B | 29.1% | huggingface.co | Official leaderboard | ||
| 146 | izhx/udever-bloom-7b1 open weights | 29.1% | huggingface.co | Official leaderboard | ||
| 147 | avsolatorio/GIST-large-Embedding-v0 open weights, 0.3B | 29.1% | huggingface.co | Official leaderboard | ||
| 148 | cointegrated/rubert-tiny2 open weights, 0B | 29% | huggingface.co | Official leaderboard | ||
| 149 | Mihaiii/Venusaur open weights, 0B | 29% | huggingface.co | Official leaderboard | ||
| 150 | Mihaiii/gte-micro open weights, 0B | 29% | huggingface.co | Official leaderboard | ||
| 151 | WhereIsAI/UAE-Large-V1 open weights, 0.3B | 29% | huggingface.co | Official leaderboard | ||
| 152 | BAAI/bge-base-en-v1.5 open weights, 0.1B | 28.9% | huggingface.co | Official leaderboard | ||
| 153 | Snowflake/snowflake-arctic-embed-s open weights, 0B | 28.9% | huggingface.co | Official leaderboard | ||
| 154 | jinaai/jina-embeddings-v2-small-en open weights, 0B | 28.9% | huggingface.co | Official leaderboard | ||
| 155 | BAAI/bge-small-en-v1.5 open weights, 0B | 28.9% | huggingface.co | Official leaderboard | ||
| 156 | brahmairesearch/slx-v0.1 open weights, 0B | 28.9% | huggingface.co | Official leaderboard | ||
| 157 | sentence-transformers/all-MiniLM-L6-v2 open weights, 0B | 28.9% | huggingface.co | Official leaderboard | ||
| 158 | sentence-transformers/all-mpnet-base-v2 open weights, 0.1B | 28.8% | huggingface.co | Official leaderboard | ||
| 159 | izhx/udever-bloom-3b open weights, 3B | 28.8% | huggingface.co | Official leaderboard | ||
| 160 | thenlper/gte-small open weights, 0B | 28.8% | huggingface.co | Official leaderboard | ||
| 161 | Snowflake/snowflake-arctic-embed-m-v1.5 open weights, 0.1B | 28.8% | huggingface.co | Official leaderboard | ||
| 162 | Mihaiii/Wartortle open weights, 0B | 28.8% | huggingface.co | Official leaderboard | ||
| 163 | manu/sentence_croissant_alpha_v0.2 open weights, 1.3B | 28.7% | huggingface.co | Official leaderboard | ||
| 164 | Mihaiii/Ivysaur open weights, 0B | 28.7% | huggingface.co | Official leaderboard | ||
| 165 | izhx/udever-bloom-1b1 open weights, 1.1B | 28.6% | huggingface.co | Official leaderboard | ||
| 166 | Mihaiii/Squirtle open weights, 0B | 28.6% | huggingface.co | Official leaderboard | ||
| 167 | izhx/udever-bloom-560m open weights, 0.6B | 28.5% | huggingface.co | Official leaderboard | ||
| 168 | Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka open weights, 0.1B | 28.5% | huggingface.co | Official leaderboard | ||
| 169 | Snowflake/snowflake-arctic-embed-xs open weights, 0B | 28.4% | huggingface.co | Official leaderboard | ||
| 170 | avsolatorio/GIST-all-MiniLM-L6-v2 open weights, 0B | 28.3% | huggingface.co | Official leaderboard | ||
| 171 | ibm-granite/granite-embedding-30m-english open weights, 0B | 28.3% | huggingface.co | Official leaderboard | ||
| 172 | intfloat/e5-small-v2 open weights, 0B | 28.3% | huggingface.co | Official leaderboard | ||
| 173 | abhinand/MedEmbed-small-v0.1 open weights, 0B | 28.3% | huggingface.co | Official leaderboard | ||
| 174 | ai-forever/sbert_large_mt_nlu_ru open weights, 0.4B | 28.3% | huggingface.co | Official leaderboard | ||
| 175 | sergeyzh/rubert-tiny-turbo open weights, 0B | 28.3% | huggingface.co | Official leaderboard | ||
| 176 | manu/sentence_croissant_alpha_v0.3 open weights, 1.3B | 28.3% | huggingface.co | Official leaderboard | ||
| 177 | sentence-transformers/all-MiniLM-L12-v2 open weights, 0B | 28.1% | huggingface.co | Official leaderboard | ||
| 178 | ibm-granite/granite-embedding-125m-english open weights, 0.1B | 28% | huggingface.co | Official leaderboard | ||
| 179 | Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka open weights, 0.2B | 28% | huggingface.co | Official leaderboard | ||
| 180 | Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet open weights, 0.1B | 27.6% | huggingface.co | Official leaderboard | ||
| 181 | malenia1/ternary-weight-embedding open weights, 0.1B | 27.5% | huggingface.co | Official leaderboard | ||
| 182 | minishlab/potion-base-8M open weights, 0B | 27.3% | huggingface.co | Official leaderboard | ||
| 183 | minishlab/potion-base-2M open weights, 0B | 27.1% | huggingface.co | Official leaderboard | ||
| 184 | minishlab/M2V_base_glove open weights, 0.1B | 26.9% | huggingface.co | Official leaderboard | ||
| 185 | minishlab/potion-base-4M open weights, 0B | 26.9% | huggingface.co | Official leaderboard | ||
| 186 | minishlab/M2V_base_glove_subword open weights, 0.1B | 26.9% | huggingface.co | Official leaderboard | ||
| 187 | ai-forever/sbert_large_nlu_ru open weights, 0.4B | 26.8% | huggingface.co | Official leaderboard | ||
| 188 | consciousAI/cai-lunaris-text-embeddings open weights, 0.3B | 26.8% | huggingface.co | Official leaderboard | ||
| 189 | jinaai/jina-embeddings-v2-base-en open weights, 0.1B | 26.8% | huggingface.co | Official leaderboard | ||
| 190 | sentence-transformers/static-similarity-mrl-multilingual-v1 open weights, 0.1B | 26.7% | huggingface.co | Official leaderboard | ||
| 191 | jinaai/jina-embedding-b-en-v1 open weights, 0.1B | 26.1% | huggingface.co | Official leaderboard | ||
| 192 | minishlab/M2V_base_output open weights, 0B | 25.9% | huggingface.co | Official leaderboard | ||
| 193 | jinaai/jina-embedding-s-en-v1 open weights, 0B | 25.7% | huggingface.co | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 59.9% | Jun 2025 | huggingface.co | Official leaderboard | |
| 2 | | 59.6% | Jun 2025 | huggingface.co | Official leaderboard | |
| 3 | | 45.8% | Mar 2026 | ai.google.dev | Official leaderboard | |
| 4 | | 41.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 5 | | 38.7% | 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: MTEB’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: MTEB’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 Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
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 benchmark’s paper (193 entries).
- Pass Models scored 5 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 59.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on MTEB: 5 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.