IndicLID benchmark
Models identify Indian languages in native scripts and in Romanized text.
MTEB scores the task using accuracy.
- Made by
- Yash Madhani and colleagues
- Released
- May 2023
- Kind
- Reading in many languages
- Level
- School
- Scored
- Checked automatically
- Answers
- Prediction
- Tasks
- All public
- Measures
- Capability
- Language
- Many languages
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Many languages
- Last model added
- October 2026
- Scores refreshed
- November 20th, 2026
Bytedance/Seed1.6-embedding-1215 tops the leaderboard with 99.3% Qwen3-Embedding-8B leads with 95.1% 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 95.1%. 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
- 99.3 Bytedance/Seed1.6-embedding-1215
- 99.3 microsoft/harrier-oss-v1-27b
- 99.1 tencent/KaLM-Embedding-Gemma3-12B-2511
- 98.5 microsoft/harrier-oss-v1-0.6b
- 98.2 microsoft/harrier-oss-v1-270m
- 98.0 jinaai/jina-embeddings-v5-omni-small
- 98.0 jinaai/jina-embeddings-v5-text-small
- 97.6 jinaai/jina-embeddings-v5-omni-nano
- 97.6 jinaai/jina-embeddings-v5-text-nano
- 95.5 nvidia/llama-embed-nemotron-8b
- 95.3 Jaume/gemma-2b-embeddings
- 95.1
Qwen3-Embedding-8B
- 95.0
Qwen3-Embedding-4B
- 94.9 codefuse-ai/F2LLM-v2-14B
- 93.1
Qwen3-Embedding-0.6B
- 93.1 codefuse-ai/F2LLM-v2-8B
- 91.4 izhx/udever-bloom-7b1
- 91.2 izhx/udever-bloom-3b
- 91.0 codefuse-ai/F2LLM-v2-4B
- 90.1 Haon-Chen/speed-embedding-7b-instruct
- 88.9 izhx/udever-bloom-1b1
- 88.7 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse
- 87.8 BidirLM/BidirLM-1.7B-Embedding
- 87.7 google/gemini-embedding-001
- 87.3 BidirLM/BidirLM-Omni-2.5B-Embedding
- 87.3 codefuse-ai/F2LLM-v2-160M
- 86.8 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised
- 86.3 intfloat/multilingual-e5-large-instruct
- 86.2 nvidia/NV-Embed-v2
- 85.8 BidirLM/BidirLM-1B-Embedding
- 85.5 codefuse-ai/F2LLM-v2-330M
- 85.5
EmbeddingGemma 2
- 85.4 codefuse-ai/F2LLM-v2-0.6B
- 84.6 nvidia/NV-Embed-v1
- 84.2 codefuse-ai/F2LLM-v2-1.7B
- 82.9 BidirLM/BidirLM-270M-Embedding
- 82.7 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826
- 82.2 ai-sage/Giga-Embeddings-instruct-3B-0826
- 82.2 hotchpotch/bekko-static-embedding-v1
- 82.1 BidirLM/BidirLM-0.6B-Embedding
- 81.6 GritLM/GritLM-7B
- 81.4 ICT-TIME-and-Querit/BOOM_4B_v1
- 80.9 GritLM/GritLM-8x7B
- 80.8 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1
- 80.4 codefuse-ai/F2LLM-v2-80M
- 80.1 izhx/udever-bloom-560m
- 79.7 Lajavaness/bilingual-embedding-large
- 79.6 Alibaba-NLP/gte-Qwen1.5-7B-instruct
- 78.8 sdadas/mmlw-e5-base
- 78.8 bigscience/sgpt-bloom-7b1-msmarco
- 77.8 Salesforce/SFR-Embedding-2_R
- 77.2 Linq-AI-Research/Linq-Embed-Mistral
- 75.6 openai/text-embedding-3-small
- 74.5 intfloat/e5-mistral-7b-instruct
- 74.0 Salesforce/SFR-Embedding-Mistral
- 72.2 ibm-granite/granite-embedding-125m-english
- 71.9 deepvk/deberta-v1-base
- 71.7 Mira190/Euler-Legal-Embedding-V1
- 71.6 sdadas/mmlw-e5-large
- 70.7 nomic-ai/nomic-embed-text-v2-moe
- 70.4 mongodb/voyage-large-2-instruct
- 67.7 HIT-TMG/KaLM-embedding-multilingual-mini-v1
- 64.6 NovaSearch/jasper_en_vision_language_v1
- 64.5 ai-forever/ru-en-RoSBERTa
- 61.4 cnmoro/static-nomic-384-pten-v2-st
- 60.2 ibm-granite/granite-embedding-30m-english
- 58.9 sentence-transformers/static-similarity-mrl-multilingual-v1
- 57.3 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1
- 55.4 NovaSearch/stella_en_1.5B_v5
- 55.2 Sailesh97/Hinvec
- 54.9 ibm-granite/granite-embedding-97m-multilingual-r2
- 54.4 Snowflake/snowflake-arctic-embed-m-v2.0
- 54.3 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet
- 54.1 openai/text-embedding-3-large
- 53.2 sdadas/mmlw-e5-small
- 53.1 intfloat/e5-large-v2
- 52.1 jinaai/jina-embeddings-v2-small-en
- 51.6 dwzhu/e5-base-4k
- 51.5 deepvk/USER-base
- 51.5 minishlab/M2V_base_glove_subword
- 51.1 minishlab/M2V_base_output
- 50.3 intfloat/e5-base-v2
- 49.6 Cohere/Cohere-embed-english-v3.0
- 49.5 nomic-ai/nomic-embed-text-v1-unsupervised
- 49.1 nomic-ai/nomic-embed-text-v1-ablated
- 48.7 brahmairesearch/slx-v0.1
- 48.7 sentence-transformers/all-MiniLM-L6-v2
- 48.1 sentence-transformers/all-MiniLM-L12-v2
- 48.0 nomic-ai/nomic-embed-text-v1
- 47.9 Snowflake/snowflake-arctic-embed-m-v1.5
- 47.9 nomic-ai/nomic-embed-text-v1.5
- 47.5 Mihaiii/Venusaur
- 47.3 intfloat/e5-small-v2
- 47.1 sentence-transformers/all-mpnet-base-v2
- 46.9 minishlab/potion-base-8M
- 46.6 google/embeddinggemma-300m
- 46.5 intfloat/e5-large
- 46.4 Cohere/Cohere-embed-english-light-v3.0
- 46.3 Snowflake/snowflake-arctic-embed-m-long
- 46.0 manu/sentence_croissant_alpha_v0.3
- 45.8 manu/sentence_croissant_alpha_v0.2
- 45.8 Snowflake/snowflake-arctic-embed-m
- 45.8 minishlab/potion-base-4M
- 45.4 consciousAI/cai-lunaris-text-embeddings
- 45.4 zeroentropy/zembed-1
- 44.9 Mihaiii/gte-micro-v4
- 44.6 Mihaiii/Squirtle
- 44.6 Mihaiii/Bulbasaur
- 44.5 avsolatorio/GIST-large-Embedding-v0
- 44.3 Snowflake/snowflake-arctic-embed-l
- 44.0 minishlab/potion-base-2M
- 43.9 avsolatorio/NoInstruct-small-Embedding-v0
- 43.8 mixedbread-ai/mxbai-embed-large-v1
- 43.5 google/text-multilingual-embedding-002
- 43.5 avsolatorio/GIST-all-MiniLM-L6-v2
- 43.4 WhereIsAI/UAE-Large-V1
- 43.3 manu/sentence_croissant_alpha_v0.4
- 43.1 intfloat/e5-base
- 42.8 Mihaiii/Ivysaur
- 42.7 BAAI/bge-base-en-v1.5
- 42.7 minishlab/potion-multilingual-128M
- 42.3 Mihaiii/Wartortle
- 41.5 intfloat/e5-small
- 41.1 infgrad/stella-base-en-v2
- 40.9 NovaSearch/stella_en_400M_v5
- 40.5 abhinand/MedEmbed-small-v0.1
- 40.5 BAAI/bge-small-en-v1.5
- 40.4 avsolatorio/GIST-Embedding-v0
- 40.3 Snowflake/snowflake-arctic-embed-xs
- 40.2 Snowflake/snowflake-arctic-embed-s
- 38.7 mongodb/voyage-multilingual-2
- 38.6 avsolatorio/GIST-small-Embedding-v0
- 38.2 Alibaba-NLP/gte-Qwen2-1.5B-instruct
- 37.5 thenlper/gte-large
- 37.3 Mihaiii/gte-micro
- 37.2 thenlper/gte-base
- 37.0 BAAI/bge-large-en-v1.5
- 36.8 thenlper/gte-small
- 35.1 deepfile/embedder-100p
- 35.0 ibm-granite/granite-embedding-107m-multilingual
- 34.0 intfloat/multilingual-e5-small
- 33.0 Snowflake/snowflake-arctic-embed-l-v2.0
- 32.8 Hum-Works/lodestone-base-4096-v1
- 32.8 Lajavaness/bilingual-embedding-base
- 32.3 mongodb/voyage-3-lite
- 32.1 shibing624/text2vec-base-multilingual
- 31.3 jinaai/jina-embeddings-v2-base-en
- 30.8 Octen/Octen-Embedding-8B
- 29.1 ibm-granite/granite-embedding-311m-multilingual-r2
- 28.9 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet
- 28.6 mongodb/voyage-3.5
- 27.7 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
- 27.3 DeepPavlov/rubert-base-cased-sentence
- 27.1 ibm-granite/granite-embedding-278m-multilingual
- 26.8 OrdalieTech/Solon-embeddings-large-0.1
- 26.3 Lajavaness/bilingual-embedding-small
- 25.7 DeepPavlov/rubert-base-cased
- 25.2 mongodb/voyage-3
- 24.3 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka
- 22.9 intfloat/multilingual-e5-base
- 22.5 Cohere/Cohere-embed-multilingual-v3.0
- 22.4 Gameselo/STS-multilingual-mpnet-base-v2
- 22.4 Kowshik24/bangla-sentence-transformer-ft-matryoshka-paraphrase-multilingual-mpnet-base-v2
- 20.8 BAAI/bge-m3
- 20.3 manu/bge-m3-custom-fr
- 20.3 intfloat/multilingual-e5-large
- 19.5 Cohere/Cohere-embed-multilingual-light-v3.0
- 19.2 sentence-transformers/LaBSE
- 19.1 Alibaba-NLP/gte-Qwen2-7B-instruct
- 18.0 jinaai/jina-embeddings-v3
- 17.9 sentence-transformers/paraphrase-multilingual-mpnet-base-v2
- 17.8 minishlab/M2V_base_glove
- 16.8 omarelshehy/arabic-english-sts-matryoshka
- 16.6 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka
- 15.9 DeepPavlov/distilrubert-small-cased-conversational
- 15.5 sdadas/mmlw-roberta-large
- 15.2 cointegrated/rubert-tiny
- 15.1 sdadas/mmlw-roberta-base
- 14.8 hotchpotch/bekko-embedding-v1-a8m
- 14.6 ai-forever/sbert_large_nlu_ru
- 14.4 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka
- 14.2 ai-forever/sbert_large_mt_nlu_ru
- 14.2 sergeyzh/LaBSE-ru-turbo
- 14.2 aari1995/German_Semantic_STS_V2
- 14.0 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka
- 13.3 malenia1/ternary-weight-embedding
- 13.2 cointegrated/LaBSE-en-ru
- 13.1 hotchpotch/bekko-embedding-v1-a25m
- 13.1 deepvk/USER-bge-m3
- 12.9 silma-ai/silma-embeddding-matryoshka-v0.1
- 12.4 Alibaba-NLP/gte-multilingual-base
- 10.9 cointegrated/rubert-tiny2
- 10.7 sergeyzh/rubert-tiny-turbo
- 10.3 jinaai/jina-embedding-b-en-v1
- 8.31 jinaai/jina-embedding-s-en-v1
- 1 Bytedance/Seed1.6-embedding-1215 99.3%
- 2 microsoft/harrier-oss-v1-27b 99.3%
- 3 tencent/KaLM-Embedding-Gemma3-12B-2511 99.1%
- 4 microsoft/harrier-oss-v1-0.6b 98.5%
- 5 microsoft/harrier-oss-v1-270m 98.2%
- 6 jinaai/jina-embeddings-v5-omni-small 98%
- 7 jinaai/jina-embeddings-v5-text-small 98%
- 8 jinaai/jina-embeddings-v5-omni-nano 97.6%
- 9 jinaai/jina-embeddings-v5-text-nano 97.6%
- 10 nvidia/llama-embed-nemotron-8b 95.5%
- 11 Jaume/gemma-2b-embeddings 95.3%
- 12
Qwen3-Embedding-8B 95.1%
- 13
Qwen3-Embedding-4B 95%
- 14 codefuse-ai/F2LLM-v2-14B 94.9%
- 15
Qwen3-Embedding-0.6B 93.1%
- 16 codefuse-ai/F2LLM-v2-8B 93.1%
- 17 izhx/udever-bloom-7b1 91.4%
- 18 izhx/udever-bloom-3b 91.2%
- 19 codefuse-ai/F2LLM-v2-4B 91%
- 20 Haon-Chen/speed-embedding-7b-instruct 90.1%
- 21 izhx/udever-bloom-1b1 88.9%
- 22 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse 88.7%
- 23 BidirLM/BidirLM-1.7B-Embedding 87.8%
- 24 google/gemini-embedding-001 87.7%
- 25 BidirLM/BidirLM-Omni-2.5B-Embedding 87.3%
- 26 codefuse-ai/F2LLM-v2-160M 87.3%
- 27 McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised 86.8%
- 28 intfloat/multilingual-e5-large-instruct 86.3%
- 29 nvidia/NV-Embed-v2 86.2%
- 30 BidirLM/BidirLM-1B-Embedding 85.8%
- 31 codefuse-ai/F2LLM-v2-330M 85.5%
- 32
EmbeddingGemma 2 85.5%
- 33 codefuse-ai/F2LLM-v2-0.6B 85.4%
- 34 nvidia/NV-Embed-v1 84.6%
- 35 codefuse-ai/F2LLM-v2-1.7B 84.2%
- 36 BidirLM/BidirLM-270M-Embedding 82.9%
- 37 ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 82.7%
- 38 ai-sage/Giga-Embeddings-instruct-3B-0826 82.2%
- 39 hotchpotch/bekko-static-embedding-v1 82.2%
- 40 BidirLM/BidirLM-0.6B-Embedding 82.1%
- 41 GritLM/GritLM-7B 81.6%
- 42 ICT-TIME-and-Querit/BOOM_4B_v1 81.4%
- 43 GritLM/GritLM-8x7B 80.9%
- 44 ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 80.8%
- 45 codefuse-ai/F2LLM-v2-80M 80.4%
- 46 izhx/udever-bloom-560m 80.1%
- 47 Lajavaness/bilingual-embedding-large 79.7%
- 48 Alibaba-NLP/gte-Qwen1.5-7B-instruct 79.6%
- 49 sdadas/mmlw-e5-base 78.8%
- 50 bigscience/sgpt-bloom-7b1-msmarco 78.8%
- 51 Salesforce/SFR-Embedding-2_R 77.8%
- 52 Linq-AI-Research/Linq-Embed-Mistral 77.2%
- 53 openai/text-embedding-3-small 75.6%
- 54 intfloat/e5-mistral-7b-instruct 74.5%
- 55 Salesforce/SFR-Embedding-Mistral 74%
- 56 ibm-granite/granite-embedding-125m-english 72.2%
- 57 deepvk/deberta-v1-base 71.9%
- 58 Mira190/Euler-Legal-Embedding-V1 71.7%
- 59 sdadas/mmlw-e5-large 71.6%
- 60 nomic-ai/nomic-embed-text-v2-moe 70.7%
- 61 mongodb/voyage-large-2-instruct 70.4%
- 62 HIT-TMG/KaLM-embedding-multilingual-mini-v1 67.7%
- 63 NovaSearch/jasper_en_vision_language_v1 64.6%
- 64 ai-forever/ru-en-RoSBERTa 64.5%
- 65 cnmoro/static-nomic-384-pten-v2-st 61.4%
- 66 ibm-granite/granite-embedding-30m-english 60.2%
- 67 sentence-transformers/static-similarity-mrl-multilingual-v1 58.9%
- 68 HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 57.3%
- 69 NovaSearch/stella_en_1.5B_v5 55.4%
- 70 Sailesh97/Hinvec 55.2%
- 71 ibm-granite/granite-embedding-97m-multilingual-r2 54.9%
- 72 Snowflake/snowflake-arctic-embed-m-v2.0 54.4%
- 73 Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet 54.3%
- 74 openai/text-embedding-3-large 54.1%
- 75 sdadas/mmlw-e5-small 53.2%
- 76 intfloat/e5-large-v2 53.1%
- 77 jinaai/jina-embeddings-v2-small-en 52.1%
- 78 dwzhu/e5-base-4k 51.6%
- 79 deepvk/USER-base 51.5%
- 80 minishlab/M2V_base_glove_subword 51.5%
- 81 minishlab/M2V_base_output 51.1%
- 82 intfloat/e5-base-v2 50.3%
- 83 Cohere/Cohere-embed-english-v3.0 49.6%
- 84 nomic-ai/nomic-embed-text-v1-unsupervised 49.5%
- 85 nomic-ai/nomic-embed-text-v1-ablated 49.1%
- 86 brahmairesearch/slx-v0.1 48.7%
- 87 sentence-transformers/all-MiniLM-L6-v2 48.7%
- 88 sentence-transformers/all-MiniLM-L12-v2 48.1%
- 89 nomic-ai/nomic-embed-text-v1 48%
- 90 Snowflake/snowflake-arctic-embed-m-v1.5 47.9%
- 91 nomic-ai/nomic-embed-text-v1.5 47.9%
- 92 Mihaiii/Venusaur 47.5%
- 93 intfloat/e5-small-v2 47.3%
- 94 sentence-transformers/all-mpnet-base-v2 47.1%
- 95 minishlab/potion-base-8M 46.9%
- 96 google/embeddinggemma-300m 46.6%
- 97 intfloat/e5-large 46.5%
- 98 Cohere/Cohere-embed-english-light-v3.0 46.4%
- 99 Snowflake/snowflake-arctic-embed-m-long 46.3%
- 100 manu/sentence_croissant_alpha_v0.3 46%
- 101 manu/sentence_croissant_alpha_v0.2 45.8%
- 102 Snowflake/snowflake-arctic-embed-m 45.8%
- 103 minishlab/potion-base-4M 45.8%
- 104 consciousAI/cai-lunaris-text-embeddings 45.4%
- 105 zeroentropy/zembed-1 45.4%
- 106 Mihaiii/gte-micro-v4 44.9%
- 107 Mihaiii/Squirtle 44.6%
- 108 Mihaiii/Bulbasaur 44.6%
- 109 avsolatorio/GIST-large-Embedding-v0 44.5%
- 110 Snowflake/snowflake-arctic-embed-l 44.3%
- 111 minishlab/potion-base-2M 44%
- 112 avsolatorio/NoInstruct-small-Embedding-v0 43.9%
- 113 mixedbread-ai/mxbai-embed-large-v1 43.8%
- 114 google/text-multilingual-embedding-002 43.5%
- 115 avsolatorio/GIST-all-MiniLM-L6-v2 43.5%
- 116 WhereIsAI/UAE-Large-V1 43.4%
- 117 manu/sentence_croissant_alpha_v0.4 43.3%
- 118 intfloat/e5-base 43.1%
- 119 Mihaiii/Ivysaur 42.8%
- 120 BAAI/bge-base-en-v1.5 42.7%
- 121 minishlab/potion-multilingual-128M 42.7%
- 122 Mihaiii/Wartortle 42.3%
- 123 intfloat/e5-small 41.5%
- 124 infgrad/stella-base-en-v2 41.1%
- 125 NovaSearch/stella_en_400M_v5 40.9%
- 126 abhinand/MedEmbed-small-v0.1 40.5%
- 127 BAAI/bge-small-en-v1.5 40.5%
- 128 avsolatorio/GIST-Embedding-v0 40.4%
- 129 Snowflake/snowflake-arctic-embed-xs 40.3%
- 130 Snowflake/snowflake-arctic-embed-s 40.2%
- 131 mongodb/voyage-multilingual-2 38.7%
- 132 avsolatorio/GIST-small-Embedding-v0 38.6%
- 133 Alibaba-NLP/gte-Qwen2-1.5B-instruct 38.2%
- 134 thenlper/gte-large 37.5%
- 135 Mihaiii/gte-micro 37.3%
- 136 thenlper/gte-base 37.2%
- 137 BAAI/bge-large-en-v1.5 37%
- 138 thenlper/gte-small 36.8%
- 139 deepfile/embedder-100p 35.1%
- 140 ibm-granite/granite-embedding-107m-multilingual 35%
- 141 intfloat/multilingual-e5-small 34%
- 142 Snowflake/snowflake-arctic-embed-l-v2.0 33%
- 143 Hum-Works/lodestone-base-4096-v1 32.8%
- 144 Lajavaness/bilingual-embedding-base 32.8%
- 145 mongodb/voyage-3-lite 32.3%
- 146 shibing624/text2vec-base-multilingual 32.1%
- 147 jinaai/jina-embeddings-v2-base-en 31.3%
- 148 Octen/Octen-Embedding-8B 30.8%
- 149 ibm-granite/granite-embedding-311m-multilingual-r2 29.1%
- 150 Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet 28.9%
- 151 mongodb/voyage-3.5 28.6%
- 152 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 27.7%
- 153 DeepPavlov/rubert-base-cased-sentence 27.3%
- 154 ibm-granite/granite-embedding-278m-multilingual 27.1%
- 155 OrdalieTech/Solon-embeddings-large-0.1 26.8%
- 156 Lajavaness/bilingual-embedding-small 26.3%
- 157 DeepPavlov/rubert-base-cased 25.7%
- 158 mongodb/voyage-3 25.2%
- 159 Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka 24.3%
- 160 intfloat/multilingual-e5-base 22.9%
- 161 Cohere/Cohere-embed-multilingual-v3.0 22.5%
- 162 Gameselo/STS-multilingual-mpnet-base-v2 22.4%
- 163 Kowshik24/bangla-sentence-transformer-ft-matryoshka-paraphrase-multilingual-mpnet-base-v2 22.4%
- 164 BAAI/bge-m3 20.8%
- 165 manu/bge-m3-custom-fr 20.3%
- 166 intfloat/multilingual-e5-large 20.3%
- 167 Cohere/Cohere-embed-multilingual-light-v3.0 19.5%
- 168 sentence-transformers/LaBSE 19.2%
- 169 Alibaba-NLP/gte-Qwen2-7B-instruct 19.1%
- 170 jinaai/jina-embeddings-v3 18%
- 171 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 17.9%
- 172 minishlab/M2V_base_glove 17.8%
- 173 omarelshehy/arabic-english-sts-matryoshka 16.8%
- 174 Omartificial-Intelligence-Space/Arabic-labse-Matryoshka 16.6%
- 175 DeepPavlov/distilrubert-small-cased-conversational 15.9%
- 176 sdadas/mmlw-roberta-large 15.5%
- 177 cointegrated/rubert-tiny 15.2%
- 178 sdadas/mmlw-roberta-base 15.1%
- 179 hotchpotch/bekko-embedding-v1-a8m 14.8%
- 180 ai-forever/sbert_large_nlu_ru 14.6%
- 181 Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka 14.4%
- 182 ai-forever/sbert_large_mt_nlu_ru 14.2%
- 183 sergeyzh/LaBSE-ru-turbo 14.2%
- 184 aari1995/German_Semantic_STS_V2 14.2%
- 185 Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka 14%
- 186 malenia1/ternary-weight-embedding 13.3%
- 187 cointegrated/LaBSE-en-ru 13.2%
- 188 hotchpotch/bekko-embedding-v1-a25m 13.1%
- 189 deepvk/USER-bge-m3 13.1%
- 190 silma-ai/silma-embeddding-matryoshka-v0.1 12.9%
- 191 Alibaba-NLP/gte-multilingual-base 12.4%
- 192 cointegrated/rubert-tiny2 10.9%
- 193 sergeyzh/rubert-tiny-turbo 10.7%
- 194 jinaai/jina-embedding-b-en-v1 10.3%
- 195 jinaai/jina-embedding-s-en-v1 8.31%
Source: MTEB’s leaderboard
- 1
Qwen3-Embedding-8B 95.1%
- 2
Qwen3-Embedding-4B 95%
- 3
Qwen3-Embedding-0.6B 93.1%
- 4
EmbeddingGemma 2 85.5%
Source: MTEB’s leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | Bytedance/Seed1.6-embedding-1215 closed weights | 99.3% | console.volcengine.com | Official leaderboard | ||
| 2 | microsoft/harrier-oss-v1-27b open weights, 27B | 99.3% | huggingface.co | Official leaderboard | ||
| 3 | tencent/KaLM-Embedding-Gemma3-12B-2511 open weights, 11.8B | 99.1% | huggingface.co | Official leaderboard | ||
| 4 | microsoft/harrier-oss-v1-0.6b open weights, 0.6B | 98.5% | huggingface.co | Official leaderboard | ||
| 5 | microsoft/harrier-oss-v1-270m open weights, 0.3B | 98.2% | huggingface.co | Official leaderboard | ||
| 6 | jinaai/jina-embeddings-v5-omni-small open weights, 1.6B | 98% | huggingface.co | Official leaderboard | ||
| 7 | jinaai/jina-embeddings-v5-text-small open weights, 0.6B | 98% | huggingface.co | Official leaderboard | ||
| 8 | jinaai/jina-embeddings-v5-omni-nano open weights, 1B | 97.6% | huggingface.co | Official leaderboard | ||
| 9 | jinaai/jina-embeddings-v5-text-nano open weights, 0.2B | 97.6% | huggingface.co | Official leaderboard | ||
| 10 | nvidia/llama-embed-nemotron-8b open weights, 7.5B | 95.5% | huggingface.co | Official leaderboard | ||
| 11 | Jaume/gemma-2b-embeddings open weights, 2.5B | 95.3% | huggingface.co | Official leaderboard | ||
| 12 | | 95.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 13 | | 95% | Jun 2025 | huggingface.co | Official leaderboard | |
| 14 | codefuse-ai/F2LLM-v2-14B open weights, 14B | 94.9% | huggingface.co | Official leaderboard | ||
| 15 | | 93.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 16 | codefuse-ai/F2LLM-v2-8B open weights, 7.6B | 93.1% | huggingface.co | Official leaderboard | ||
| 17 | izhx/udever-bloom-7b1 open weights | 91.4% | huggingface.co | Official leaderboard | ||
| 18 | izhx/udever-bloom-3b open weights, 3B | 91.2% | huggingface.co | Official leaderboard | ||
| 19 | codefuse-ai/F2LLM-v2-4B open weights, 4B | 91% | huggingface.co | Official leaderboard | ||
| 20 | Haon-Chen/speed-embedding-7b-instruct open weights, 7.1B | 90.1% | huggingface.co | Official leaderboard | ||
| 21 | izhx/udever-bloom-1b1 open weights, 1.1B | 88.9% | huggingface.co | Official leaderboard | ||
| 22 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-unsup-simcse open weights, 7.2B | 88.7% | huggingface.co | Official leaderboard | ||
| 23 | BidirLM/BidirLM-1.7B-Embedding open weights, 1.7B | 87.8% | huggingface.co | Official leaderboard | ||
| 24 | google/gemini-embedding-001 closed weights | 87.7% | ai.google.dev | Official leaderboard | ||
| 25 | BidirLM/BidirLM-Omni-2.5B-Embedding open weights, 2.4B | 87.3% | huggingface.co | Official leaderboard | ||
| 26 | codefuse-ai/F2LLM-v2-160M open weights, 0.2B | 87.3% | huggingface.co | Official leaderboard | ||
| 27 | McGill-NLP/LLM2Vec-Mistral-7B-Instruct-v2-mntp-supervised open weights, 7.2B | 86.8% | huggingface.co | Official leaderboard | ||
| 28 | intfloat/multilingual-e5-large-instruct open weights, 0.6B | 86.3% | huggingface.co | Official leaderboard | ||
| 29 | nvidia/NV-Embed-v2 open weights, 7.9B | 86.2% | huggingface.co | Official leaderboard | ||
| 30 | BidirLM/BidirLM-1B-Embedding open weights, 1B | 85.8% | huggingface.co | Official leaderboard | ||
| 31 | codefuse-ai/F2LLM-v2-330M open weights, 0.3B | 85.5% | huggingface.co | Official leaderboard | ||
| 32 | | 85.5% | Oct 2026 | huggingface.co | Official leaderboard | |
| 33 | codefuse-ai/F2LLM-v2-0.6B open weights, 0.6B | 85.4% | huggingface.co | Official leaderboard | ||
| 34 | nvidia/NV-Embed-v1 open weights, 7.9B | 84.6% | huggingface.co | Official leaderboard | ||
| 35 | codefuse-ai/F2LLM-v2-1.7B open weights, 1.7B | 84.2% | huggingface.co | Official leaderboard | ||
| 36 | BidirLM/BidirLM-270M-Embedding open weights, 0.3B | 82.9% | huggingface.co | Official leaderboard | ||
| 37 | ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826 open weights, 10.5B | 82.7% | huggingface.co | Official leaderboard | ||
| 38 | ai-sage/Giga-Embeddings-instruct-3B-0826 open weights, 3.2B | 82.2% | huggingface.co | Official leaderboard | ||
| 39 | hotchpotch/bekko-static-embedding-v1 open weights, 0.2B | 82.2% | huggingface.co | Official leaderboard | ||
| 40 | BidirLM/BidirLM-0.6B-Embedding open weights, 0.6B | 82.1% | huggingface.co | Official leaderboard | ||
| 41 | GritLM/GritLM-7B open weights, 7.2B | 81.6% | huggingface.co | Official leaderboard | ||
| 42 | ICT-TIME-and-Querit/BOOM_4B_v1 open weights, 4B | 81.4% | huggingface.co | Official leaderboard | ||
| 43 | GritLM/GritLM-8x7B open weights, 57.9B | 80.9% | huggingface.co | Official leaderboard | ||
| 44 | ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1 open weights, 4B | 80.8% | huggingface.co | Official leaderboard | ||
| 45 | codefuse-ai/F2LLM-v2-80M open weights, 0.1B | 80.4% | huggingface.co | Official leaderboard | ||
| 46 | izhx/udever-bloom-560m open weights, 0.6B | 80.1% | huggingface.co | Official leaderboard | ||
| 47 | Lajavaness/bilingual-embedding-large open weights, 0.6B | 79.7% | huggingface.co | Official leaderboard | ||
| 48 | Alibaba-NLP/gte-Qwen1.5-7B-instruct open weights, 7.1B | 79.6% | huggingface.co | Official leaderboard | ||
| 49 | sdadas/mmlw-e5-base open weights, 0.3B | 78.8% | huggingface.co | Official leaderboard | ||
| 50 | bigscience/sgpt-bloom-7b1-msmarco open weights, 7.1B | 78.8% | huggingface.co | Official leaderboard | ||
| 51 | Salesforce/SFR-Embedding-2_R open weights, 7.1B | 77.8% | huggingface.co | Official leaderboard | ||
| 52 | Linq-AI-Research/Linq-Embed-Mistral open weights, 7.1B | 77.2% | huggingface.co | Official leaderboard | ||
| 53 | openai/text-embedding-3-small closed weights | 75.6% | openai.com | Official leaderboard | ||
| 54 | intfloat/e5-mistral-7b-instruct open weights, 7.1B | 74.5% | huggingface.co | Official leaderboard | ||
| 55 | Salesforce/SFR-Embedding-Mistral open weights, 7.1B | 74% | huggingface.co | Official leaderboard | ||
| 56 | ibm-granite/granite-embedding-125m-english open weights, 0.1B | 72.2% | huggingface.co | Official leaderboard | ||
| 57 | deepvk/deberta-v1-base open weights, 0.1B | 71.9% | huggingface.co | Official leaderboard | ||
| 58 | Mira190/Euler-Legal-Embedding-V1 open weights, 7.6B | 71.7% | huggingface.co | Official leaderboard | ||
| 59 | sdadas/mmlw-e5-large open weights, 0.6B | 71.6% | huggingface.co | Official leaderboard | ||
| 60 | nomic-ai/nomic-embed-text-v2-moe open weights, 0.5B | 70.7% | huggingface.co | Official leaderboard | ||
| 61 | mongodb/voyage-large-2-instruct closed weights | 70.4% | blog.voyageai.com | Official leaderboard | ||
| 62 | HIT-TMG/KaLM-embedding-multilingual-mini-v1 open weights, 0.5B | 67.7% | huggingface.co | Official leaderboard | ||
| 63 | NovaSearch/jasper_en_vision_language_v1 open weights, 1.6B | 64.6% | huggingface.co | Official leaderboard | ||
| 64 | ai-forever/ru-en-RoSBERTa open weights, 0.4B | 64.5% | huggingface.co | Official leaderboard | ||
| 65 | cnmoro/static-nomic-384-pten-v2-st open weights, 0.1B | 61.4% | huggingface.co | Official leaderboard | ||
| 66 | ibm-granite/granite-embedding-30m-english open weights, 0B | 60.2% | huggingface.co | Official leaderboard | ||
| 67 | sentence-transformers/static-similarity-mrl-multilingual-v1 open weights, 0.1B | 58.9% | huggingface.co | Official leaderboard | ||
| 68 | HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1 open weights, 0.5B | 57.3% | huggingface.co | Official leaderboard | ||
| 69 | NovaSearch/stella_en_1.5B_v5 open weights, 1.5B | 55.4% | huggingface.co | Official leaderboard | ||
| 70 | Sailesh97/Hinvec open weights, 0.9B | 55.2% | huggingface.co | Official leaderboard | ||
| 71 | ibm-granite/granite-embedding-97m-multilingual-r2 open weights, 0.1B | 54.9% | huggingface.co | Official leaderboard | ||
| 72 | Snowflake/snowflake-arctic-embed-m-v2.0 open weights, 0.3B | 54.4% | huggingface.co | Official leaderboard | ||
| 73 | Omartificial-Intelligence-Space/Arabic-mpnet-base-all-nli-triplet open weights, 0.1B | 54.3% | huggingface.co | Official leaderboard | ||
| 74 | openai/text-embedding-3-large closed weights | 54.1% | openai.com | Official leaderboard | ||
| 75 | sdadas/mmlw-e5-small open weights, 0.1B | 53.2% | huggingface.co | Official leaderboard | ||
| 76 | intfloat/e5-large-v2 open weights, 0.3B | 53.1% | huggingface.co | Official leaderboard | ||
| 77 | jinaai/jina-embeddings-v2-small-en open weights, 0B | 52.1% | huggingface.co | Official leaderboard | ||
| 78 | dwzhu/e5-base-4k open weights, 0.1B | 51.6% | huggingface.co | Official leaderboard | ||
| 79 | deepvk/USER-base open weights, 0.4B | 51.5% | huggingface.co | Official leaderboard | ||
| 80 | minishlab/M2V_base_glove_subword open weights, 0.1B | 51.5% | huggingface.co | Official leaderboard | ||
| 81 | minishlab/M2V_base_output open weights, 0B | 51.1% | huggingface.co | Official leaderboard | ||
| 82 | intfloat/e5-base-v2 open weights, 0.1B | 50.3% | huggingface.co | Official leaderboard | ||
| 83 | Cohere/Cohere-embed-english-v3.0 closed weights | 49.6% | cohere.com | Official leaderboard | ||
| 84 | nomic-ai/nomic-embed-text-v1-unsupervised open weights, 0.1B | 49.5% | huggingface.co | Official leaderboard | ||
| 85 | nomic-ai/nomic-embed-text-v1-ablated open weights, 0.1B | 49.1% | huggingface.co | Official leaderboard | ||
| 86 | brahmairesearch/slx-v0.1 open weights, 0B | 48.7% | huggingface.co | Official leaderboard | ||
| 87 | sentence-transformers/all-MiniLM-L6-v2 open weights, 0B | 48.7% | huggingface.co | Official leaderboard | ||
| 88 | sentence-transformers/all-MiniLM-L12-v2 open weights, 0B | 48.1% | huggingface.co | Official leaderboard | ||
| 89 | nomic-ai/nomic-embed-text-v1 open weights, 0.1B | 48% | huggingface.co | Official leaderboard | ||
| 90 | Snowflake/snowflake-arctic-embed-m-v1.5 open weights, 0.1B | 47.9% | huggingface.co | Official leaderboard | ||
| 91 | nomic-ai/nomic-embed-text-v1.5 open weights, 0.1B | 47.9% | huggingface.co | Official leaderboard | ||
| 92 | Mihaiii/Venusaur open weights, 0B | 47.5% | huggingface.co | Official leaderboard | ||
| 93 | intfloat/e5-small-v2 open weights, 0B | 47.3% | huggingface.co | Official leaderboard | ||
| 94 | sentence-transformers/all-mpnet-base-v2 open weights, 0.1B | 47.1% | huggingface.co | Official leaderboard | ||
| 95 | minishlab/potion-base-8M open weights, 0B | 46.9% | huggingface.co | Official leaderboard | ||
| 96 | google/embeddinggemma-300m open weights, 0.3B | 46.6% | ai.google.dev | Official leaderboard | ||
| 97 | intfloat/e5-large open weights, 0.3B | 46.5% | huggingface.co | Official leaderboard | ||
| 98 | Cohere/Cohere-embed-english-light-v3.0 closed weights | 46.4% | cohere.com | Official leaderboard | ||
| 99 | Snowflake/snowflake-arctic-embed-m-long open weights, 0.1B | 46.3% | huggingface.co | Official leaderboard | ||
| 100 | manu/sentence_croissant_alpha_v0.3 open weights, 1.3B | 46% | huggingface.co | Official leaderboard | ||
| 101 | manu/sentence_croissant_alpha_v0.2 open weights, 1.3B | 45.8% | huggingface.co | Official leaderboard | ||
| 102 | Snowflake/snowflake-arctic-embed-m open weights, 0.1B | 45.8% | huggingface.co | Official leaderboard | ||
| 103 | minishlab/potion-base-4M open weights, 0B | 45.8% | huggingface.co | Official leaderboard | ||
| 104 | consciousAI/cai-lunaris-text-embeddings open weights, 0.3B | 45.4% | huggingface.co | Official leaderboard | ||
| 105 | zeroentropy/zembed-1 open weights, 4B | 45.4% | huggingface.co | Official leaderboard | ||
| 106 | Mihaiii/gte-micro-v4 open weights, 0B | 44.9% | huggingface.co | Official leaderboard | ||
| 107 | Mihaiii/Squirtle open weights, 0B | 44.6% | huggingface.co | Official leaderboard | ||
| 108 | Mihaiii/Bulbasaur open weights, 0B | 44.6% | huggingface.co | Official leaderboard | ||
| 109 | avsolatorio/GIST-large-Embedding-v0 open weights, 0.3B | 44.5% | huggingface.co | Official leaderboard | ||
| 110 | Snowflake/snowflake-arctic-embed-l open weights, 0.3B | 44.3% | huggingface.co | Official leaderboard | ||
| 111 | minishlab/potion-base-2M open weights, 0B | 44% | huggingface.co | Official leaderboard | ||
| 112 | avsolatorio/NoInstruct-small-Embedding-v0 open weights, 0B | 43.9% | huggingface.co | Official leaderboard | ||
| 113 | mixedbread-ai/mxbai-embed-large-v1 open weights, 0.3B | 43.8% | huggingface.co | Official leaderboard | ||
| 114 | google/text-multilingual-embedding-002 closed weights | 43.5% | cloud.google.com | Official leaderboard | ||
| 115 | avsolatorio/GIST-all-MiniLM-L6-v2 open weights, 0B | 43.5% | huggingface.co | Official leaderboard | ||
| 116 | WhereIsAI/UAE-Large-V1 open weights, 0.3B | 43.4% | huggingface.co | Official leaderboard | ||
| 117 | manu/sentence_croissant_alpha_v0.4 open weights, 1.3B | 43.3% | huggingface.co | Official leaderboard | ||
| 118 | intfloat/e5-base open weights, 0.1B | 43.1% | huggingface.co | Official leaderboard | ||
| 119 | Mihaiii/Ivysaur open weights, 0B | 42.8% | huggingface.co | Official leaderboard | ||
| 120 | BAAI/bge-base-en-v1.5 open weights, 0.1B | 42.7% | huggingface.co | Official leaderboard | ||
| 121 | minishlab/potion-multilingual-128M open weights, 0.1B | 42.7% | huggingface.co | Official leaderboard | ||
| 122 | Mihaiii/Wartortle open weights, 0B | 42.3% | huggingface.co | Official leaderboard | ||
| 123 | intfloat/e5-small open weights, 0B | 41.5% | huggingface.co | Official leaderboard | ||
| 124 | infgrad/stella-base-en-v2 open weights, 0.1B | 41.1% | huggingface.co | Official leaderboard | ||
| 125 | NovaSearch/stella_en_400M_v5 open weights, 0.4B | 40.9% | huggingface.co | Official leaderboard | ||
| 126 | abhinand/MedEmbed-small-v0.1 open weights, 0B | 40.5% | huggingface.co | Official leaderboard | ||
| 127 | BAAI/bge-small-en-v1.5 open weights, 0B | 40.5% | huggingface.co | Official leaderboard | ||
| 128 | avsolatorio/GIST-Embedding-v0 open weights, 0.1B | 40.4% | huggingface.co | Official leaderboard | ||
| 129 | Snowflake/snowflake-arctic-embed-xs open weights, 0B | 40.3% | huggingface.co | Official leaderboard | ||
| 130 | Snowflake/snowflake-arctic-embed-s open weights, 0B | 40.2% | huggingface.co | Official leaderboard | ||
| 131 | mongodb/voyage-multilingual-2 closed weights | 38.7% | blog.voyageai.com | Official leaderboard | ||
| 132 | avsolatorio/GIST-small-Embedding-v0 open weights, 0B | 38.6% | huggingface.co | Official leaderboard | ||
| 133 | Alibaba-NLP/gte-Qwen2-1.5B-instruct open weights, 1.5B | 38.2% | huggingface.co | Official leaderboard | ||
| 134 | thenlper/gte-large open weights, 0.3B | 37.5% | huggingface.co | Official leaderboard | ||
| 135 | Mihaiii/gte-micro open weights, 0B | 37.3% | huggingface.co | Official leaderboard | ||
| 136 | thenlper/gte-base open weights, 0.1B | 37.2% | huggingface.co | Official leaderboard | ||
| 137 | BAAI/bge-large-en-v1.5 open weights, 0.3B | 37% | huggingface.co | Official leaderboard | ||
| 138 | thenlper/gte-small open weights, 0B | 36.8% | huggingface.co | Official leaderboard | ||
| 139 | deepfile/embedder-100p open weights, 0.3B | 35.1% | huggingface.co | Official leaderboard | ||
| 140 | ibm-granite/granite-embedding-107m-multilingual open weights, 0.1B | 35% | huggingface.co | Official leaderboard | ||
| 141 | intfloat/multilingual-e5-small open weights, 0.1B | 34% | huggingface.co | Official leaderboard | ||
| 142 | Snowflake/snowflake-arctic-embed-l-v2.0 open weights, 0.6B | 33% | huggingface.co | Official leaderboard | ||
| 143 | Hum-Works/lodestone-base-4096-v1 open weights, 0.1B | 32.8% | huggingface.co | Official leaderboard | ||
| 144 | Lajavaness/bilingual-embedding-base open weights, 0.3B | 32.8% | huggingface.co | Official leaderboard | ||
| 145 | mongodb/voyage-3-lite closed weights | 32.3% | blog.voyageai.com | Official leaderboard | ||
| 146 | shibing624/text2vec-base-multilingual open weights, 0.1B | 32.1% | huggingface.co | Official leaderboard | ||
| 147 | jinaai/jina-embeddings-v2-base-en open weights, 0.1B | 31.3% | huggingface.co | Official leaderboard | ||
| 148 | Octen/Octen-Embedding-8B open weights, 7.6B | 30.8% | huggingface.co | Official leaderboard | ||
| 149 | ibm-granite/granite-embedding-311m-multilingual-r2 open weights, 0.3B | 29.1% | huggingface.co | Official leaderboard | ||
| 150 | Omartificial-Intelligence-Space/Arabic-MiniLM-L12-v2-all-nli-triplet open weights, 0.1B | 28.9% | huggingface.co | Official leaderboard | ||
| 151 | mongodb/voyage-3.5 closed weights | 28.6% | blog.voyageai.com | Official leaderboard | ||
| 152 | sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 open weights, 0.1B | 27.7% | huggingface.co | Official leaderboard | ||
| 153 | DeepPavlov/rubert-base-cased-sentence open weights, 0.1B | 27.3% | huggingface.co | Official leaderboard | ||
| 154 | ibm-granite/granite-embedding-278m-multilingual open weights, 0.3B | 27.1% | huggingface.co | Official leaderboard | ||
| 155 | OrdalieTech/Solon-embeddings-large-0.1 open weights, 0.6B | 26.8% | huggingface.co | Official leaderboard | ||
| 156 | Lajavaness/bilingual-embedding-small open weights, 0.1B | 26.3% | huggingface.co | Official leaderboard | ||
| 157 | DeepPavlov/rubert-base-cased open weights, 1.3B | 25.7% | huggingface.co | Official leaderboard | ||
| 158 | mongodb/voyage-3 closed weights | 25.2% | blog.voyageai.com | Official leaderboard | ||
| 159 | Omartificial-Intelligence-Space/Arabic-all-nli-triplet-Matryoshka open weights, 0.3B | 24.3% | huggingface.co | Official leaderboard | ||
| 160 | intfloat/multilingual-e5-base open weights, 0.3B | 22.9% | huggingface.co | Official leaderboard | ||
| 161 | Cohere/Cohere-embed-multilingual-v3.0 closed weights | 22.5% | cohere.com | Official leaderboard | ||
| 162 | Gameselo/STS-multilingual-mpnet-base-v2 open weights, 0.3B | 22.4% | huggingface.co | Official leaderboard | ||
| 163 | Kowshik24/bangla-sentence-transformer-ft-matryoshka-paraphrase-multilingual-mpnet-base-v2 open weights, 0.3B | 22.4% | huggingface.co | Official leaderboard | ||
| 164 | BAAI/bge-m3 open weights, 0.6B | 20.8% | huggingface.co | Official leaderboard | ||
| 165 | manu/bge-m3-custom-fr open weights, 0.6B | 20.3% | huggingface.co | Official leaderboard | ||
| 166 | intfloat/multilingual-e5-large open weights, 0.6B | 20.3% | huggingface.co | Official leaderboard | ||
| 167 | Cohere/Cohere-embed-multilingual-light-v3.0 closed weights | 19.5% | cohere.com | Official leaderboard | ||
| 168 | sentence-transformers/LaBSE open weights, 0.5B | 19.2% | huggingface.co | Official leaderboard | ||
| 169 | Alibaba-NLP/gte-Qwen2-7B-instruct open weights, 7.1B | 19.1% | huggingface.co | Official leaderboard | ||
| 170 | jinaai/jina-embeddings-v3 open weights, 0.6B | 18% | huggingface.co | Official leaderboard | ||
| 171 | sentence-transformers/paraphrase-multilingual-mpnet-base-v2 open weights, 0.3B | 17.9% | huggingface.co | Official leaderboard | ||
| 172 | minishlab/M2V_base_glove open weights, 0.1B | 17.8% | huggingface.co | Official leaderboard | ||
| 173 | omarelshehy/arabic-english-sts-matryoshka open weights, 0.6B | 16.8% | huggingface.co | Official leaderboard | ||
| 174 | Omartificial-Intelligence-Space/Arabic-labse-Matryoshka open weights, 0.5B | 16.6% | huggingface.co | Official leaderboard | ||
| 175 | DeepPavlov/distilrubert-small-cased-conversational open weights, 0.1B | 15.9% | huggingface.co | Official leaderboard | ||
| 176 | sdadas/mmlw-roberta-large open weights, 0.4B | 15.5% | huggingface.co | Official leaderboard | ||
| 177 | cointegrated/rubert-tiny open weights, 0B | 15.2% | huggingface.co | Official leaderboard | ||
| 178 | sdadas/mmlw-roberta-base open weights, 0.1B | 15.1% | huggingface.co | Official leaderboard | ||
| 179 | hotchpotch/bekko-embedding-v1-a8m open weights, 0.1B | 14.8% | huggingface.co | Official leaderboard | ||
| 180 | ai-forever/sbert_large_nlu_ru open weights, 0.4B | 14.6% | huggingface.co | Official leaderboard | ||
| 181 | Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka open weights, 0.2B | 14.4% | huggingface.co | Official leaderboard | ||
| 182 | ai-forever/sbert_large_mt_nlu_ru open weights, 0.4B | 14.2% | huggingface.co | Official leaderboard | ||
| 183 | sergeyzh/LaBSE-ru-turbo open weights, 0.1B | 14.2% | huggingface.co | Official leaderboard | ||
| 184 | aari1995/German_Semantic_STS_V2 open weights, 0.3B | 14.2% | huggingface.co | Official leaderboard | ||
| 185 | Omartificial-Intelligence-Space/Arabert-all-nli-triplet-Matryoshka open weights, 0.1B | 14% | huggingface.co | Official leaderboard | ||
| 186 | malenia1/ternary-weight-embedding open weights, 0.1B | 13.3% | huggingface.co | Official leaderboard | ||
| 187 | cointegrated/LaBSE-en-ru open weights, 0.1B | 13.2% | huggingface.co | Official leaderboard | ||
| 188 | hotchpotch/bekko-embedding-v1-a25m open weights, 0.1B | 13.1% | huggingface.co | Official leaderboard | ||
| 189 | deepvk/USER-bge-m3 open weights, 0.4B | 13.1% | huggingface.co | Official leaderboard | ||
| 190 | silma-ai/silma-embeddding-matryoshka-v0.1 open weights, 0.1B | 12.9% | huggingface.co | Official leaderboard | ||
| 191 | Alibaba-NLP/gte-multilingual-base open weights, 0.3B | 12.4% | huggingface.co | Official leaderboard | ||
| 192 | cointegrated/rubert-tiny2 open weights, 0B | 10.9% | huggingface.co | Official leaderboard | ||
| 193 | sergeyzh/rubert-tiny-turbo open weights, 0B | 10.7% | huggingface.co | Official leaderboard | ||
| 194 | jinaai/jina-embedding-b-en-v1 open weights, 0.1B | 10.3% | huggingface.co | Official leaderboard | ||
| 195 | jinaai/jina-embedding-s-en-v1 open weights, 0B | 8.31% | huggingface.co | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 95.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 2 | | 95% | Jun 2025 | huggingface.co | Official leaderboard | |
| 3 | | 93.1% | Jun 2025 | huggingface.co | Official leaderboard | |
| 4 | | 85.5% | 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
Qwen3-Embedding-8B on the official leaderboard passed 85% in June 2025
That was 25 months after its release. By the same age, 45% of the benchmarks from 2023 had passed 85%. 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 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 (195 entries).
- Pass Models scored 4 models we track, plus 191 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 Saturated. The best is 95.1%, past 85% since June 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on MTEB: 4 rows for models we track and 191 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.