Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
mteb (Massive Text Embedding Benchmark)
[go: Go Back, main page]

mteb, check out our documentation.

\n
\n\t\n\t\t\n\n\n\n\n\t\t\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\t
Overview
πŸ“ˆ LeaderboardThe interactive leaderboard of the benchmark
Get Started.
πŸƒ Get StartedOverview of how to use mteb
πŸ€– Defining ModelsHow to use existing model and define custom ones
πŸ“‹ Selecting tasksHow to select tasks, benchmarks, splits etc.
🏭 Running EvaluationHow to run the evaluations, including cache management, speeding up evaluations etc.
πŸ“Š Loading ResultsHow to load and work with existing model results
Overview.
πŸ“‹ TasksOverview of available tasks
πŸ“ BenchmarksOverview of available benchmarks
πŸ€– ModelsOverview of available Models
Contributing
πŸ€– Adding a modelHow to submit a model to MTEB and to the leaderboard
πŸ‘©β€πŸ’» Adding a datasetHow to add a new task/dataset to MTEB
πŸ‘©β€πŸ’» Adding a benchmarkHow to add a new benchmark to MTEB and to the leaderboard
🀝 ContributingHow to contribute to MTEB and set it up for development
\n
\n","classNames":"hf-sanitized hf-sanitized-s7akt_Bd9IiMG-tQjNtl6"},"users":[{"_id":"624bf78fbe275a1baf4dbd91","avatarUrl":"/avatars/be91dd24ea9f42a55c68d75e5a8681d7.svg","isPro":false,"fullname":"LoΓ―c Magne","user":"loicmagne","type":"user"},{"_id":"5ff8c9f4b2035d9a81a859f7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1652134289581-5ff8c9f4b2035d9a81a859f7.jpeg","isPro":false,"fullname":"Nouamane Tazi","user":"nouamanetazi","type":"user"},{"_id":"5f1eb362eec0ad2a071ad6e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f1eb362eec0ad2a071ad6e2/IXMYkYKuTwn6kBdWnQeeY.png","isPro":false,"fullname":"Niklas Muennighoff","user":"Muennighoff","type":"user"},{"_id":"5eff4688ff69163f6f59e66c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1596792577829-5eff4688ff69163f6f59e66c.jpeg","isPro":false,"fullname":"Nils Reimers","user":"nreimers","type":"user"},{"_id":"6317233cc92fd6fee317e030","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6317233cc92fd6fee317e030/cJHSvvimr1kqgQfHOjO5n.png","isPro":false,"fullname":"Tom Aarsen","user":"tomaarsen","type":"user"},{"_id":"642c54a5b09c70b36de03071","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/642c54a5b09c70b36de03071/EwQyQmust01dgkdRMtYFt.jpeg","isPro":true,"fullname":"rasdani","user":"rasdani","type":"user"},{"_id":"62610f8040e04009e81047e9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62610f8040e04009e81047e9/iqGTYB7OMmS0jkFEB7cEF.jpeg","isPro":false,"fullname":"Imene Kerboua","user":"imenelydiaker","type":"user"},{"_id":"5ff5943752c26e9bc240bada","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/Exyzf3C_gJ2KdsL4K5_cq.png","isPro":false,"fullname":"Kenneth C. Enevoldsen","user":"KennethEnevoldsen","type":"user"},{"_id":"647895fe60410176020a5c58","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/647895fe60410176020a5c58/56k-9Tln9S5epsb-e6Zr6.jpeg","isPro":false,"fullname":"Wissam Siblini","user":"Wissam42","type":"user"},{"_id":"629e4ad01fec321a2d2abfa3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/629e4ad01fec321a2d2abfa3/LwMzgQr9RH5arKOBNR043.jpeg","isPro":false,"fullname":"Shuang Li","user":"Shuang59","type":"user"},{"_id":"648ba0d68e7f7a927675d4a3","avatarUrl":"/avatars/d82ced93656e03d60c8b55010694f908.svg","isPro":false,"fullname":"Hong Liu","user":"hongliu9903","type":"user"},{"_id":"6362d9712691058b19de1ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6362d9712691058b19de1ba4/Hdqj5aGrFJJbF7oUSzoIh.jpeg","isPro":true,"fullname":"Orion Weller","user":"orionweller","type":"user"},{"_id":"6610d8304a55d90baf3c238c","avatarUrl":"/avatars/2317948c0b69336d164cccda5bfdee52.svg","isPro":false,"fullname":"Thakur","user":"Sakshamrzt","type":"user"},{"_id":"63434eb76f59b79da07dbddf","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63434eb76f59b79da07dbddf/BEwmVjqPNYlqmutXG0G6e.jpeg","isPro":false,"fullname":"Sara Hooker","user":"sarahooker","type":"user"},{"_id":"5fc53305fa6eef7667a4d691","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1623089209793-5fc53305fa6eef7667a4d691.jpeg","isPro":false,"fullname":"Manan Dey","user":"manandey","type":"user"},{"_id":"62696cd3d1ac0cde59280dcf","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1656940964610-62696cd3d1ac0cde59280dcf.jpeg","isPro":false,"fullname":"MΓ‘rton Kardos","user":"kardosdrur","type":"user"},{"_id":"62645f88c39850dc093d6105","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1650745211725-noauth.png","isPro":false,"fullname":"Mohammed Hamdy","user":"mmhamdy","type":"user"},{"_id":"662cc845616128914a3c9817","avatarUrl":"/avatars/cd186de00f28bb866abc1ab6c4465663.svg","isPro":false,"fullname":"DomKrz","user":"dokato","type":"user"},{"_id":"6400f2ed568dbe30c9161e47","avatarUrl":"/avatars/c55938df5bce82b5d96e592a1ec36a8b.svg","isPro":false,"fullname":"Weijia Shi","user":"swj0419","type":"user"},{"_id":"5f3801ab7e583543386217ac","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f3801ab7e583543386217ac/4xMdDV1gws7nxCJrU321H.jpeg","isPro":false,"fullname":"Aaron Chibb","user":"aari1995","type":"user"},{"_id":"624734dc4c731bb6bfab8af7","avatarUrl":"/avatars/6b250b58710a3287b85e4733c1824558.svg","isPro":false,"fullname":"Siva Reddy","user":"sivareddyg","type":"user"},{"_id":"61981af5d420757268e195ac","avatarUrl":"/avatars/8b59aaf33447224f83d497425fd7ea8f.svg","isPro":false,"fullname":"Vaibhav Adlakha","user":"vaibhavad","type":"user"},{"_id":"60e1bc418479fac0bd1daa0e","avatarUrl":"/avatars/61f4f8ac0714aae7d5cbb7d4e1038020.svg","isPro":false,"fullname":"Andrianos Michail","user":"Andrianos","type":"user"},{"_id":"64cc0e80a257a3212c0c4b24","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64cc0e80a257a3212c0c4b24/wqs6WZN8-3OQthcnQXgN7.png","isPro":false,"fullname":"Isaac Chung","user":"isaacchung","type":"user"},{"_id":"6454e5ac273f64983024ba5d","avatarUrl":"/avatars/ca6f861ed830a79f7a1eba04ebe84afc.svg","isPro":false,"fullname":"Vatolin Alexey","user":"vatolinalex","type":"user"},{"_id":"64d2fce8129a210e569e0c76","avatarUrl":"/avatars/a79a832dc3a46ece1b9e542369fc4888.svg","isPro":false,"fullname":"Dawei Zhu","user":"dwzhu","type":"user"},{"_id":"63108cc834c7d77420b0fd68","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63108cc834c7d77420b0fd68/taDnqEmcI9Rhe3uzcPEE3.jpeg","isPro":false,"fullname":"Chenghao Xiao","user":"gowitheflow","type":"user"},{"_id":"61af4544d691b3aadd1f62b6","avatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","isPro":false,"fullname":"Solomatin Roman","user":"Samoed","type":"user"},{"_id":"670d9809cb84658d97d01a90","avatarUrl":"/avatars/9d3058c51b8222e25ead588919ba4b81.svg","isPro":false,"fullname":"Kyler Wang","user":"kylerywang","type":"user"},{"_id":"6671be9ff022d14aa10df864","avatarUrl":"/avatars/dd085abefa38c1604dc2ceabf472816d.svg","isPro":false,"fullname":"Adnan El Assadi","user":"AdnanElAssadi","type":"user"},{"_id":"65e7bf52ea6e08f11594f7d1","avatarUrl":"/avatars/561916d564a0d6e13465b9f3d4672124.svg","isPro":false,"fullname":"Mina Parham","user":"Mina76","type":"user"},{"_id":"670d8a232d412a30dfa28ac2","avatarUrl":"/avatars/6c8997140f5e864750b04d21625f86d9.svg","isPro":true,"fullname":"Silky Singh","user":"silky1708","type":"user"},{"_id":"63212881cafe12f481a4e488","avatarUrl":"/avatars/91a402b11fa792566e65091143a5bc97.svg","isPro":false,"fullname":"Pengfei He","user":"hepengfe","type":"user"},{"_id":"67835e771d8713ae81c7a7ec","avatarUrl":"/avatars/ec36ae567709963ba9b727cb1f059706.svg","isPro":false,"fullname":"Sufen","user":"Sufen","type":"user"},{"_id":"66fd9e8b001816b29e0bc2e4","avatarUrl":"/avatars/190e6711cef3126c51e15276bb5ed8cd.svg","isPro":false,"fullname":"Ali Sartaz Khan","user":"alisartazkhan","type":"user"},{"_id":"63ca499104c97982831127ec","avatarUrl":"/avatars/816a962c62c805adf7789fa8e398b01e.svg","isPro":false,"fullname":"Aradhye Agarwal","user":"aradhye","type":"user"},{"_id":"67c0c512f217797d4dbe8ace","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/5qEVZY8ky_caKcwAZsteS.png","isPro":false,"fullname":"Shikhar Shiromani","user":"sshiromani1729","type":"user"},{"_id":"64c1094bb85ee9e4223402f5","avatarUrl":"/avatars/5c598ac8472ab27c3d69bd4c5a4c9e2d.svg","isPro":false,"fullname":"SmileHu","user":"SmileXing","type":"user"},{"_id":"64784057403cd7ae4b7897c1","avatarUrl":"/avatars/88d5de26b847b25bba27e99a5fb44e58.svg","isPro":false,"fullname":"Jay Jung","user":"diffunity","type":"user"},{"_id":"66b0c560acd41a0acb530ea9","avatarUrl":"/avatars/b3e0d58d8e21a68039a82ecdf9b9f766.svg","isPro":false,"fullname":"Peichun Hua","user":"pchua","type":"user"},{"_id":"64d02f09a2e7f9ff61035013","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64d02f09a2e7f9ff61035013/V9HK7GVnpBwenNz8SSyus.jpeg","isPro":false,"fullname":"Yevhen Kostiuk","user":"yevhenkost","type":"user"}],"userCount":41,"collections":[{"slug":"mteb/mmteb-67b74a586236bc839971e8cd","title":"MMTEB","description":"A collection of items telated the the MMTEB release","gating":false,"lastUpdated":"2026-02-20T16:56:16.593Z","owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"items":[{"_id":"67b74a586236bc839971e8ce","position":0,"type":"paper","id":"2502.13595","title":"MMTEB: Massive Multilingual Text Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2502.13595.png","upvotes":44,"publishedAt":"2025-02-19T10:13:43.000Z","isUpvotedByUser":false},{"_id":"67b84a1f342b6cfb48a90202","position":1,"type":"space","author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"colorFrom":"blue","colorTo":"indigo","createdAt":"2022-09-29T11:29:23.000Z","emoji":"πŸ₯‡","id":"mteb/leaderboard","lastModified":"2026-01-20T21:57:52.000Z","likes":7047,"pinned":true,"private":false,"sdk":"docker","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-upgrade","requested":"cpu-upgrade"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"mteb-leaderboard.hf.space","stage":"READY"}],"sha":"5132ddc89608299ca2f183fc13b60e1fafd2de7d"},"shortDescription":"Embedding Leaderboard","title":"MTEB Leaderboard","isLikedByUser":false,"ai_short_description":"Explore multilingual text embedding benchmark results","ai_category":"Text Analysis","trendingScore":17,"tags":["docker","leaderboard","region:us"],"featured":false},{"_id":"69988f2a7d7ae20f45f3ca74","position":2,"type":"dataset","author":"mteb","downloads":5863,"gated":false,"id":"mteb/BornholmBitextMining","lastModified":"2025-09-10T11:22:32.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":6785,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"69988f2a3ccabf2d2404f0e9","position":3,"type":"dataset","author":"davidstap","downloads":5063,"gated":false,"id":"davidstap/biblenlp-corpus-mmteb","lastModified":"2024-04-26T14:09:22.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":1743371,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["json"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":3,"isLikedByUser":false,"isBenchmark":false}],"position":0,"theme":"green","private":false,"shareUrl":"https://hf.co/collections/mteb/mmteb","upvotes":4,"isUpvotedByUser":false},{"slug":"mteb/maeb-699887357d7ae20f45f35148","title":"MAEB","description":"MAEB is a comprehensive audio benchmark with 30 tasks spanning both audio-only and audio-text cross-modal evaluation. Tasks span 7 task types: retr...","gating":false,"lastUpdated":"2026-02-20T16:56:16.588Z","owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"items":[{"_id":"6998882b7d7ae20f45f35827","position":0,"type":"paper","id":"2602.16008","title":"MAEB: Massive Audio Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2602.16008.png","upvotes":16,"publishedAt":"2026-02-17T21:00:51.000Z","isUpvotedByUser":false},{"_id":"69988735640f064c400cefe7","position":1,"type":"dataset","author":"mteb","downloads":45,"gated":false,"id":"mteb/Clotho","lastModified":"2026-01-20T08:22:54.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":1045,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"699887353ccabf2d240477ac","position":2,"type":"dataset","author":"mteb","downloads":294,"gated":false,"id":"mteb/common_voice_21_0_mini","lastModified":"2025-12-30T16:14:59.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":77534,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"699887356ab28243867a3436","position":3,"type":"dataset","author":"mteb","downloads":522,"gated":false,"id":"mteb/fleurs","lastModified":"2026-01-17T11:30:54.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":384060,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false}],"position":1,"theme":"orange","private":false,"shareUrl":"https://hf.co/collections/mteb/maeb","upvotes":0,"isUpvotedByUser":false},{"slug":"mteb/mieb-699891dc5390e02222d56929","title":"MIEB","description":"MIEB(Multilingual) is a comprehensive image embeddings benchmark, spanning 10 task types, covering 130 tasks and a total of 39 languages.\n In ad...","gating":false,"lastUpdated":"2026-02-20T16:56:16.591Z","owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"items":[{"_id":"6998920bdf79a976c55a39af","position":0,"type":"paper","id":"2504.10471","title":"MIEB: Massive Image Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2504.10471.png","upvotes":21,"publishedAt":"2025-04-14T17:54:28.000Z","isUpvotedByUser":false},{"_id":"699891dd0cab71e3d90f8645","position":1,"type":"dataset","author":"mteb","downloads":98,"gated":false,"id":"mteb/birdsnap","lastModified":"2024-10-26T10:35:08.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":17858,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"699891dd3ccabf2d24052dc3","position":2,"type":"dataset","author":"mteb","downloads":122,"gated":false,"id":"mteb/Caltech101","lastModified":"2025-06-05T14:25:02.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":9144,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["image"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"699891dddf79a976c55a351d","position":3,"type":"dataset","author":"mteb","downloads":79,"gated":false,"id":"mteb/cifar10","lastModified":"2026-02-09T08:00:14.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":60000,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["parquet"],"modalities":["image"]},"private":false,"repoType":"dataset","likes":2,"isLikedByUser":false,"isBenchmark":false}],"position":2,"theme":"green","private":false,"shareUrl":"https://hf.co/collections/mteb/mieb","upvotes":0,"isUpvotedByUser":false},{"slug":"mteb/hume-699892257d7ae20f45f40ca9","title":"HUME","description":"The HUME benchmark is designed to evaluate the performance of text embedding models and humans on a comparable set of tasks. This captures areas wh...","gating":false,"lastUpdated":"2026-02-20T16:56:31.704Z","owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"items":[{"_id":"6998923a0cab71e3d90f8d6c","position":0,"type":"paper","id":"2510.10062","title":"HUME: Measuring the Human-Model Performance Gap in Text Embedding Task","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2510.10062.png","upvotes":10,"publishedAt":"2025-10-11T06:56:53.000Z","isUpvotedByUser":false},{"_id":"699892253ccabf2d24053386","position":1,"type":"dataset","author":"mteb","downloads":22,"gated":false,"id":"mteb/HUMEEmotionClassification","lastModified":"2025-09-29T09:43:30.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":16048,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"69989226d9b39c5c79717d7a","position":2,"type":"dataset","author":"mteb","downloads":26,"gated":false,"id":"mteb/HUMEToxicConversationsClassification","lastModified":"2025-09-29T10:33:57.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":8045,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"_id":"69989226c055587af218b836","position":3,"type":"dataset","author":"mteb","downloads":21,"gated":false,"id":"mteb/HUMETweetSentimentExtractionClassification","lastModified":"2025-09-29T10:49:39.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":27526,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false}],"position":3,"theme":"pink","private":false,"shareUrl":"https://hf.co/collections/mteb/hume","upvotes":0,"isUpvotedByUser":false},{"slug":"mteb/mteb-papers-67b84a2e10a9714460f0db4f","title":"MTEB Papers","description":"This is a collection of MTEB papers (not exhaustive).","gating":false,"lastUpdated":"2026-02-20T16:56:16.616Z","owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"items":[{"_id":"67b84a447614a7cdef8b1ca1","position":0,"type":"paper","id":"2502.13595","title":"MMTEB: Massive Multilingual Text Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2502.13595.png","upvotes":44,"publishedAt":"2025-02-19T10:13:43.000Z","isUpvotedByUser":false},{"_id":"67b84a3f00245b72c59f7efe","position":1,"type":"paper","id":"2210.07316","title":"MTEB: Massive Text Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2210.07316.png","upvotes":6,"publishedAt":"2022-10-13T19:42:08.000Z","isUpvotedByUser":false},{"_id":"67ff78c1fa998532e8f57640","position":2,"type":"paper","id":"2504.10471","title":"MIEB: Massive Image Embedding Benchmark","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2504.10471.png","upvotes":21,"publishedAt":"2025-04-14T17:54:28.000Z","isUpvotedByUser":false},{"_id":"67b84a8300245b72c59f8e18","position":3,"type":"paper","id":"2406.02396","title":"The Scandinavian Embedding Benchmarks: Comprehensive Assessment of\n Multilingual and Monolingual Text Embedding","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2406.02396.png","upvotes":0,"publishedAt":"2024-06-04T15:11:27.000Z","isUpvotedByUser":false}],"position":4,"theme":"indigo","private":false,"shareUrl":"https://hf.co/collections/mteb/mteb-papers","upvotes":2,"isUpvotedByUser":false}],"datasets":[{"author":"mteb","downloads":166777,"gated":false,"id":"mteb/results","lastModified":"2026-02-20T14:36:55.000Z","private":false,"repoType":"dataset","likes":1,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":0,"gated":false,"id":"mteb/worldqa","lastModified":"2026-02-20T12:07:55.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":3284,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":9206,"gated":false,"id":"mteb/arguana","lastModified":"2026-02-20T10:02:39.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":11486,"libraries":["datasets","pandas","polars","mlcroissant"],"formats":["json"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":1,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":0,"gated":false,"id":"mteb/MVBench","lastModified":"2026-02-20T09:25:47.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":3899,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":7,"gated":false,"id":"mteb/panda-70m","lastModified":"2026-02-19T12:41:33.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":3395,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":5,"gated":false,"id":"mteb/AudioCaps_AV","lastModified":"2026-02-19T12:04:30.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":666,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":7,"gated":false,"id":"mteb/VALOR-32K","lastModified":"2026-02-19T07:49:04.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":3491,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":6,"gated":false,"id":"mteb/AVE-Dataset","lastModified":"2026-02-19T06:27:49.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":3714,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":6,"gated":false,"id":"mteb/VGGSound_AV_RETRIEVAL","lastModified":"2026-02-19T00:27:52.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":696,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"mteb","downloads":5,"gated":false,"id":"mteb/Shot2Story20K_test","lastModified":"2026-02-18T20:49:12.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":4023,"libraries":["datasets","dask","polars","mlcroissant"],"formats":["parquet","optimized-parquet"],"modalities":["audio","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false}],"models":[{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"downloads":0,"gated":false,"id":"mteb/bm25s","availableInferenceProviders":[],"lastModified":"2026-02-20T17:10:08.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"downloads":1,"gated":false,"id":"mteb/index_wikipedia_bm25","availableInferenceProviders":[],"lastModified":"2024-07-27T15:57:22.000Z","likes":1,"private":false,"repoType":"model","isLikedByUser":false},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"downloads":2,"gated":false,"id":"mteb/index_arxiv_bm25","availableInferenceProviders":[],"lastModified":"2024-07-27T15:39:34.000Z","likes":3,"private":false,"repoType":"model","isLikedByUser":false},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"downloads":1,"gated":false,"id":"mteb/index_stackexchange_bm25","availableInferenceProviders":[],"lastModified":"2024-07-27T03:49:15.000Z","likes":1,"private":false,"repoType":"model","isLikedByUser":false}],"paperPreviews":[{"_id":"2602.16008","title":"MAEB: Massive Audio Embedding Benchmark","id":"2602.16008","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2602.16008.png"},{"_id":"2510.10062","title":"HUME: Measuring the Human-Model Performance Gap in Text Embedding Task","id":"2510.10062","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2510.10062.png"}],"spaces":[{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"colorFrom":"blue","colorTo":"indigo","createdAt":"2022-09-29T11:29:23.000Z","emoji":"πŸ₯‡","id":"mteb/leaderboard","lastModified":"2026-01-20T21:57:52.000Z","likes":7047,"pinned":true,"private":false,"sdk":"docker","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-upgrade","requested":"cpu-upgrade"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"mteb-leaderboard.hf.space","stage":"READY"}],"sha":"5132ddc89608299ca2f183fc13b60e1fafd2de7d"},"shortDescription":"Embedding Leaderboard","title":"MTEB Leaderboard","isLikedByUser":false,"ai_short_description":"Explore multilingual text embedding benchmark results","ai_category":"Text Analysis","trendingScore":17,"tags":["docker","leaderboard","region:us"],"featured":false},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"colorFrom":"blue","colorTo":"indigo","createdAt":"2025-02-04T10:13:13.000Z","emoji":"πŸ₯‡","id":"mteb/leaderboard_legacy","lastModified":"2026-02-08T18:15:59.000Z","likes":37,"pinned":false,"private":false,"sdk":"gradio","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-basic","requested":"cpu-basic"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"mteb-leaderboard-legacy.hf.space","stage":"READY"}],"sha":"92a31ad4e724352299fc495f48e5d48f3cbc585f"},"title":"MTEB Legacy Leaderboard","isLikedByUser":false,"originRepo":{"name":"mteb/leaderboard","author":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false}},"ai_short_description":"Explore and filter MTEB model benchmark results","ai_category":"Data Visualization","trendingScore":0,"tags":["gradio","leaderboard","region:us"],"featured":false},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"colorFrom":"indigo","colorTo":"blue","createdAt":"2025-09-08T08:43:31.000Z","emoji":"🏒","id":"mteb/leaderboard_dev","lastModified":"2026-01-03T15:55:34.000Z","likes":11,"pinned":false,"private":false,"sdk":"docker","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":"cpu-basic","requested":"cpu-basic"},"storage":null,"gcTimeout":172800,"replicas":{"current":1,"requested":1},"devMode":false,"domains":[{"domain":"mteb-leaderboard-dev.hf.space","stage":"READY"}],"sha":"0bf23ba3a677b035564d581f85c52ebf74afe74e"},"shortDescription":"Dedicated display for RTEB benchmark results","title":"Leaderboard Dev","isLikedByUser":false,"ai_short_description":"Compare text embedding models across languages and tasks","ai_category":"Text Analysis","trendingScore":0,"tags":["docker","region:us"],"featured":true},{"author":"mteb","authorData":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"colorFrom":"indigo","colorTo":"blue","createdAt":"2024-07-13T02:13:04.000Z","emoji":"βš”οΈ","id":"mteb/arena","lastModified":"2025-09-30T22:45:49.000Z","likes":116,"pinned":false,"private":false,"sdk":"static","repoType":"space","runtime":{"stage":"RUNNING","hardware":{"current":null,"requested":null},"storage":null,"replicas":{"requested":1,"current":1}},"title":"MTEB Arena","isLikedByUser":false,"ai_short_description":"Display MTEB Arena interface","ai_category":"Other","trendingScore":0,"tags":["static","arena","leaderboard","region:us"],"featured":false}],"buckets":[],"numBuckets":0,"numDatasets":1436,"numModels":4,"numSpaces":5,"lastOrgActivities":[{"time":"2026-02-20T17:10:08.539Z","user":"Samoed","userAvatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","org":"mteb","orgAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","type":"discussion","discussionData":{"num":4,"author":{"_id":"61af4544d691b3aadd1f62b6","avatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","fullname":"Solomatin Roman","name":"Samoed","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":45,"isUserFollowing":false},"repo":{"name":"mteb/bm25s","type":"model"},"title":"Add evaluation results for model mteb/bm25s revision 0_1_10","status":"merged","createdAt":"2026-02-20T10:00:46.000Z","isPullRequest":true,"numComments":1,"topReactions":[],"numReactionUsers":0,"pinned":false,"repoOwner":{"name":"mteb","isParticipating":true,"type":"org","isDiscussionAuthor":true}},"repoId":"mteb/bm25s","repoType":"model","eventId":"6998957022ceed64ea55b572"},{"time":"2026-02-20T16:56:26.756Z","user":"Samoed","userAvatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","orgAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","type":"collection","collection":{"id":"699892257d7ae20f45f40ca9","slug":"mteb/hume-699892257d7ae20f45f40ca9","title":"HUME","description":"The HUME benchmark is designed to evaluate the performance of text embedding models and humans on a comparable set of tasks. This captures areas wh...","lastUpdated":"2026-02-20T16:56:31.704Z","numberItems":17,"owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"theme":"pink","shareUrl":"https://hf.co/collections/mteb/hume","upvotes":0,"isUpvotedByUser":false},"org":"mteb"},{"time":"2026-02-20T16:56:08.919Z","user":"Samoed","userAvatarUrl":"/avatars/7a4067accdd1005f78c3c4adad3ee0a5.svg","orgAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","type":"collection","collection":{"id":"699892257d7ae20f45f40ca9","slug":"mteb/hume-699892257d7ae20f45f40ca9","title":"HUME","description":"The HUME benchmark is designed to evaluate the performance of text embedding models and humans on a comparable set of tasks. This captures areas wh...","lastUpdated":"2026-02-20T16:56:31.704Z","numberItems":17,"owner":{"_id":"624bfda5459c48438cc39f80","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/OrZxdlg8doDNO2TZ6Q58G.png","fullname":"Massive Text Embedding Benchmark","name":"mteb","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":916,"isUserFollowing":false},"theme":"pink","shareUrl":"https://hf.co/collections/mteb/hume","upvotes":0,"isUpvotedByUser":false},"org":"mteb"}],"acceptLanguages":["*"],"canReadRepos":false,"canReadSpaces":false,"blogPosts":[],"currentRepoPage":0,"filters":{},"paperView":false}">

AI & ML interests

Massive Text Embeddings Benchmark

Recent Activity

SamoedΒ  updated a collection 28 minutes ago
HUME
SamoedΒ  updated a collection 28 minutes ago
HUME
View all activity

MTEB is a Python framework for evaluating embeddings and retrieval systems for both text and image. MTEB covers more than 1000 languages and diverse tasks, from classics like classification and clustering to use-case specialized tasks such as legal, code, or healthcare retrieval.

You can get started using mteb, check out our documentation.

Overview
πŸ“ˆ Leaderboard The interactive leaderboard of the benchmark
Get Started.
πŸƒ Get Started Overview of how to use mteb
πŸ€– Defining Models How to use existing model and define custom ones
πŸ“‹ Selecting tasks How to select tasks, benchmarks, splits etc.
🏭 Running Evaluation How to run the evaluations, including cache management, speeding up evaluations etc.
πŸ“Š Loading Results How to load and work with existing model results
Overview.
πŸ“‹ Tasks Overview of available tasks
πŸ“ Benchmarks Overview of available benchmarks
πŸ€– Models Overview of available Models
Contributing
πŸ€– Adding a model How to submit a model to MTEB and to the leaderboard
πŸ‘©β€πŸ’» Adding a dataset How to add a new task/dataset to MTEB
πŸ‘©β€πŸ’» Adding a benchmark How to add a new benchmark to MTEB and to the leaderboard
🀝 Contributing How to contribute to MTEB and set it up for development