# MTEB(rus, v1.1) · Russian > Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/mteb%3AMTEB(rus%2C%20v1.1). When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/mteb%3AMTEB(rus%2C%20v1.1). Russian text embedding quality across classification, clustering, reranking, pair classification, retrieval, and semantic similarity. In v1.1, MIRACLRetrieval and RiaNewsRetrieval were replaced with their HardNegatives variants (v2), which include improved default prompts. Canonical URL: https://benchize.com/benchmarks/mteb%3AMTEB(rus%2C%20v1.1) Last updated: Wed, 21 Oct 2026 18:52:20 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. Russian text embedding quality across classification, clustering, reranking, pair classification, retrieval, and semantic similarity. In v1.1, MIRACLRetrieval and RiaNewsRetrieval were replaced with their HardNegatives variants (v2), which include improved default prompts. Embeddings · Models tested: 7 · Score publishers: [MTEB](https://aclanthology.org/2023.eacl-main.148/) ## Model results | # | Model | Score | | | --- | --- | --- | --- | | 1 | [Qwen3 Embedding](https://benchize.com/models/qwen3-embedding.md) | 0.71source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.7141666086956522 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.6923460555555556 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 91 ### Reported statistics experiments Not reported rank 4 scoresByTask {"CEDRClassification":0.530499,"GeoreviewClassification":0.590869,"GeoreviewClusteringP2P":0.784751,"HeadlineClassification":0.85415,"InappropriatenessClassification":0.752051,"KinopoiskClassification":0.742133,"MIRACLReranking":0.6691,"MIRACLRetrievalHardNegatives.v2":0.73363,"MassiveIntentClassification":0.8369575,"MassiveScenarioClassification":0.8863004999999999,"RUParaPhraserSTS":0.760648,"RiaNewsRetrievalHardNegatives.v2":0.84895,"RuBQReranking":0.734738,"RuBQRetrieval":0.7631,"RuReviewsClassification":0.756006,"RuSTSBenchmarkSTS":0.891558,"RuSciBenchGRNTIClassification":0.706836,"RuSciBenchGRNTIClusteringP2P":0.653119,"RuSciBenchOECDClassification":0.566064,"RuSciBenchOECDClusteringP2P":0.563635,"STS22":0.726335,"SensitiveTopicsClassification":0.361084,"TERRa":0.713318} scoresByTaskType {"Classification":0.7434852222222221,"Clustering":0.6671683333333333,"MultilabelClassification":0.4457915,"PairClassification":0.713318,"Reranking":0.701919,"Retrieval":0.7818933333333334,"STS":0.7928470000000001} trainedOnTasks ["MIRACLReranking","MIRACLRetrievalHardNegatives.v2"] ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label Qwen/Qwen3-Embedding-8B Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/3` | | | | | 2 | [Qwen3 Embedding](https://benchize.com/models/qwen3-embedding.md) | 0.7source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.7013252826086955 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.674575761904762 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 91 ### Reported statistics experiments Not reported rank 8 scoresByTask {"CEDRClassification":0.510096,"GeoreviewClassification":0.609082,"GeoreviewClusteringP2P":0.767674,"HeadlineClassification":0.835498,"InappropriatenessClassification":0.758594,"KinopoiskClassification":0.739133,"MIRACLReranking":0.65189,"MIRACLRetrievalHardNegatives.v2":0.71019,"MassiveIntentClassification":0.8268610000000001,"MassiveScenarioClassification":0.8730994999999999,"RUParaPhraserSTS":0.765856,"RiaNewsRetrievalHardNegatives.v2":0.8442,"RuBQReranking":0.722786,"RuBQRetrieval":0.73648,"RuReviewsClassification":0.759473,"RuSTSBenchmarkSTS":0.88774,"RuSciBenchGRNTIClassification":0.682031,"RuSciBenchGRNTIClusteringP2P":0.633937,"RuSciBenchOECDClassification":0.541943,"RuSciBenchOECDClusteringP2P":0.550563,"STS22":0.701387,"SensitiveTopicsClassification":0.355713,"TERRa":0.666255} scoresByTaskType {"Classification":0.7361905,"Clustering":0.6507246666666666,"MultilabelClassification":0.4329045,"PairClassification":0.666255,"Reranking":0.687338,"Retrieval":0.7636233333333333,"STS":0.7849943333333332} trainedOnTasks ["MIRACLReranking","MIRACLRetrievalHardNegatives.v2"] ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label Qwen/Qwen3-Embedding-4B Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/7` | | | | | 3 | [Bge m3](https://benchize.com/models/bge-m3.md) | 0.62source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.6162892872176834 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.6098089240158763 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 91 ### Reported statistics experiments Not reported rank 30 scoresByTask {"CEDRClassification":0.4346971307120085,"GeoreviewClassification":0.48271484375,"GeoreviewClusteringP2P":0.6309362350401001,"HeadlineClassification":0.70322265625,"InappropriatenessClassification":0.598681640625,"KinopoiskClassification":0.5823333333333334,"MIRACLReranking":0.65384,"MIRACLRetrievalHardNegatives.v2":0.70322,"MassiveIntentClassification":0.6859504999999999,"MassiveScenarioClassification":0.733839,"RUParaPhraserSTS":0.7490066815007971,"RiaNewsRetrievalHardNegatives.v2":0.84068,"RuBQReranking":0.7402919214478741,"RuBQRetrieval":0.71217,"RuReviewsClassification":0.669091796875,"RuSTSBenchmarkSTS":0.7986695688743493,"RuSciBenchGRNTIClassification":0.558056640625,"RuSciBenchGRNTIClusteringP2P":0.5083470670121341,"RuSciBenchOECDClassification":0.42568359375,"RuSciBenchOECDClusteringP2P":0.432082498945941,"STS22":0.6626491471661151,"SensitiveTopicsClassification":0.262451171875,"TERRa":0.6060381782240654} scoresByTaskType {"Classification":0.6043971116898148,"Clustering":0.523788600332725,"MultilabelClassification":0.3485741512935042,"PairClassification":0.6060381782240654,"Reranking":0.6970659607239371,"Retrieval":0.7520233333333333,"STS":0.7367751325137538} trainedOnTasks ["MIRACLReranking","MIRACLRetrievalHardNegatives.v2"] ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label BAAI/bge-m3 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/29` | | | | | 4 | [E5-Large-v2](https://benchize.com/models/e5-large-v2.md) | 0.34source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.3425063628023342 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.3405963024801973 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 86 scoresByTask {"CEDRClassification":0.3554729011689692,"GeoreviewClassification":0.320947,"GeoreviewClusteringP2P":0.276006,"HeadlineClassification":0.443848,"InappropriatenessClassification":0.537061,"KinopoiskClassification":0.372733,"MIRACLReranking":0.21609,"MIRACLRetrievalHardNegatives.v2":0.1066,"MassiveIntentClassification":0.40161782505273613,"MassiveScenarioClassification":0.4505905,"RUParaPhraserSTS":0.548945,"RiaNewsRetrievalHardNegatives.v2":0.22248,"RuBQReranking":0.47835819736633195,"RuBQRetrieval":0.21286,"RuReviewsClassification":0.479883,"RuSTSBenchmarkSTS":0.637907,"RuSciBenchGRNTIClassification":0.243311,"RuSciBenchGRNTIClusteringP2P":0.203602,"RuSciBenchOECDClassification":0.184473,"RuSciBenchOECDClusteringP2P":0.179983,"STS22":0.343741,"SensitiveTopicsClassification":0.188477,"TERRa":0.4726599208656485} scoresByTaskType {"Classification":0.38160714722808176,"Clustering":0.21986366666666668,"MultilabelClassification":0.2719749505844846,"PairClassification":0.4726599208656485,"Reranking":0.34722409868316595,"Retrieval":0.1806466666666667,"STS":0.5101976666666667} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label intfloat/e5-large-v2 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/85` | | | | | 5 | [BGE Large EN V1.5](https://benchize.com/models/bge-large-en-v1-5.md) | 0.32source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.3225534890152861 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.3328015894720617 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 96 scoresByTask {"CEDRClassification":0.3615302869287992,"GeoreviewClassification":0.286376953125,"GeoreviewClusteringP2P":0.21872509542751403,"HeadlineClassification":0.335595703125,"InappropriatenessClassification":0.518115234375,"KinopoiskClassification":0.3568,"MIRACLReranking":0.25964,"MIRACLRetrievalHardNegatives.v2":0.17327,"MassiveIntentClassification":0.3209864902279834,"MassiveScenarioClassification":0.37967799999999996,"RUParaPhraserSTS":0.49107601175437415,"RiaNewsRetrievalHardNegatives.v2":0.33746,"RuBQReranking":0.4766395487120344,"RuBQRetrieval":0.21052,"RuReviewsClassification":0.446240234375,"RuSTSBenchmarkSTS":0.603270996059513,"RuSciBenchGRNTIClassification":0.2205078125,"RuSciBenchGRNTIClusteringP2P":0.19005480904556968,"RuSciBenchOECDClassification":0.165283203125,"RuSciBenchOECDClusteringP2P":0.15984730532206398,"STS22":0.25185,"SensitiveTopicsClassification":0.179736328125,"TERRa":0.4755262351237274} scoresByTaskType {"Classification":0.3366204034281092,"Clustering":0.18954240326504923,"MultilabelClassification":0.2706333075268996,"PairClassification":0.4755262351237274,"Reranking":0.3681397743560172,"Retrieval":0.2404166666666667,"STS":0.4487323359379624} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label BAAI/bge-large-en-v1.5 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/95` | | | | | 6 | [E5-Base-v2](https://benchize.com/models/e5-base-v2.md) | 0.32source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.3162036466597676 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.31729900343053963 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 94 scoresByTask {"CEDRClassification":0.3410733262486717,"GeoreviewClassification":0.309033,"GeoreviewClusteringP2P":0.232717,"HeadlineClassification":0.417432,"InappropriatenessClassification":0.532715,"KinopoiskClassification":0.346733,"MIRACLReranking":0.16005,"MIRACLRetrievalHardNegatives.v2":0.07342,"MassiveIntentClassification":0.36346096403293204,"MassiveScenarioClassification":0.423688,"RUParaPhraserSTS":0.549563,"RiaNewsRetrievalHardNegatives.v2":0.186,"RuBQReranking":0.45350643601453516,"RuBQRetrieval":0.16225,"RuReviewsClassification":0.471289,"RuSTSBenchmarkSTS":0.643508,"RuSciBenchGRNTIClassification":0.20874,"RuSciBenchGRNTIClusteringP2P":0.182769,"RuSciBenchOECDClassification":0.164941,"RuSciBenchOECDClusteringP2P":0.161579,"STS22":0.216185,"SensitiveTopicsClassification":0.181396,"TERRa":0.49063514687851495} scoresByTaskType {"Classification":0.3597813293369925,"Clustering":0.19235499999999997,"MultilabelClassification":0.26123466312433585,"PairClassification":0.49063514687851495,"Reranking":0.30677821800726757,"Retrieval":0.14055666666666666,"STS":0.469752} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label intfloat/e5-base-v2 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/93` | | | | | 7 | [BGE Base EN V1.5](https://benchize.com/models/bge-base-en-v1-5.md) | 0.3source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.29632991759278804 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.3064628340541747 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 103 scoresByTask {"CEDRClassification":0.336238,"GeoreviewClassification":0.277393,"GeoreviewClusteringP2P":0.230877,"HeadlineClassification":0.300684,"InappropriatenessClassification":0.51626,"KinopoiskClassification":0.3558,"MIRACLReranking":0.22421,"MIRACLRetrievalHardNegatives.v2":0.11812,"MassiveIntentClassification":0.31545040622598675,"MassiveScenarioClassification":0.367614,"RUParaPhraserSTS":0.487348,"RiaNewsRetrievalHardNegatives.v2":0.26052,"RuBQReranking":0.44524547699694,"RuBQRetrieval":0.13268,"RuReviewsClassification":0.434668,"RuSTSBenchmarkSTS":0.597007,"RuSciBenchGRNTIClassification":0.173438,"RuSciBenchGRNTIClusteringP2P":0.155395,"RuSciBenchOECDClassification":0.131592,"RuSciBenchOECDClusteringP2P":0.141578,"STS22":0.161843,"SensitiveTopicsClassification":0.180469,"TERRa":0.4711582214111984} scoresByTaskType {"Classification":0.3192110451362208,"Clustering":0.17595000000000002,"MultilabelClassification":0.2583535,"PairClassification":0.4711582214111984,"Reranking":0.33472773849847,"Retrieval":0.17044,"STS":0.4153993333333333} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label BAAI/bge-base-en-v1.5 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/102` | | | | | 8 | [all-MiniLM-L6-v2](https://benchize.com/models/all-minilm-l6-v2.md) | 0.24source score | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.23662332737308422 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.23964686297992457 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 133 scoresByTask {"CEDRClassification":0.3271519659936238,"GeoreviewClassification":0.270751953125,"GeoreviewClusteringP2P":0.2024938042780367,"HeadlineClassification":0.277734375,"InappropriatenessClassification":0.517333984375,"KinopoiskClassification":0.3393333333333334,"MIRACLReranking":0.04672,"MIRACLRetrievalHardNegatives.v2":0.01185,"MassiveIntentClassification":0.2736058630445663,"MassiveScenarioClassification":0.302074,"RUParaPhraserSTS":0.43932891844400296,"RiaNewsRetrievalHardNegatives.v2":0.00738,"RuBQReranking":0.27049018085869947,"RuBQRetrieval":0.02642,"RuReviewsClassification":0.417919921875,"RuSTSBenchmarkSTS":0.5556356112936773,"RuSciBenchGRNTIClassification":0.100830078125,"RuSciBenchGRNTIClusteringP2P":0.10214186324385237,"RuSciBenchOECDClassification":0.082958984375,"RuSciBenchOECDClusteringP2P":0.09442013836007132,"STS22":0.14721380413194854,"SensitiveTopicsClassification":0.17822265625,"TERRa":0.45032509347412525} scoresByTaskType {"Classification":0.2869491659169888,"Clustering":0.13301860196065346,"MultilabelClassification":0.25268731112181186,"PairClassification":0.45032509347412525,"Reranking":0.15860509042934973,"Retrieval":0.015216666666666665,"STS":0.38072611128987627} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(rus%2C%20v1.1)/scores) Source MTEB Original model label sentence-transformers/all-MiniLM-L6-v2 Observed 2026-10-21T16:40:39.422Z Evidence SHA-256 `ac823b2b6938144aafb5350443bcb7c1f622d0aa834bf0b375d59b7ee34ed05e` Source locator `/rows/132` | | | | --- Cite as: MTEB(rus, v1.1) · Russian — Benchize, https://benchize.com/benchmarks/mteb%3AMTEB(rus%2C%20v1.1) (retrieved 2026-10-21)