# MTEB(fas, v2) · Persian

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/mteb%3AMTEB(fas%2C%20v2). When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/mteb%3AMTEB(fas%2C%20v2).

Persian text embedding quality across classification, clustering, pair classification, reranking, retrieval, semantic similarity, and summarization retrieval. In v2, large datasets were optimized for accessibility, low-quality datasets were removed, and higher-quality data was added; see the [main PR](https://github.com/embeddings-benchmark/mteb/pull/3157) for details.

Canonical URL: https://benchize.com/benchmarks/mteb%3AMTEB(fas%2C%20v2)
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.

Persian text embedding quality across classification, clustering, pair classification, reranking, retrieval, semantic similarity, and summarization retrieval. In v2, large datasets were optimized for accessibility, low-quality datasets were removed, and higher-quality data was added; see the [main PR](https://github.com/embeddings-benchmark/mteb/pull/3157) for details.

Embeddings · Models tested: 3 · Score publishers: [MTEB](https://arxiv.org/abs/2502.11571)

## Model results

| # | Model | Score |  |
| --- | --- | --- | --- |
| 1 | [Bge m3](https://benchize.com/models/bge-m3.md) | 0.66source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.6556908377594628 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.6864206419376083 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 90 ### Reported statistics experiments Not reported rank 8 scoresByTask {"ArguAna-Fa.v2":0.54025,"BeytooteClustering":0.607116,"DeepSentiPers.v2":0.667762,"DigikalamagClassification":0.860329,"DigikalamagClustering":0.395619,"FEVER-FaHardNegatives":0.64209,"FarsTail":0.731393,"FarsiParaphraseDetection":0.955728,"Farsick":0.717459,"FiQA2018-Fa.v2":0.30225,"HamshahriClustring":0.694767,"HotpotQA-FaHardNegatives":0.57622,"MIRACLReranking":0.6092,"MIRACLRetrievalHardNegatives":0.61385,"MSMARCO-FaHardNegatives":0.68471,"MassiveIntentClassification":0.6953644999999999,"MassiveScenarioClassification":0.729133,"NLPTwitterAnalysisClassification.v2":0.772556,"NLPTwitterAnalysisClustering":0.809023,"NQ-FaHardNegatives":0.50208,"NeuCLIR2023RetrievalHardNegatives":0.59298,"ParsinluEntail":0.686506,"ParsinluQueryParaphPC":0.899788,"PerShopDomainClassification":0.664646,"PerShopIntentClassification":0.898804,"PersianFoodSentimentClassification":0.830786,"PersianTextEmotion.v2":0.598117,"PersianWebDocumentRetrieval":0.44091,"QuoraRetrieval-Fa.v2":0.8031,"SAMSumFa":0.978753,"SCIDOCS-Fa.v2":0.14994,"SIDClassification.v2":0.59625,"SIDClustring":0.380031,"SciFact-Fa.v2":0.58576,"StyleClassification":0.558594,"SynPerChatbotConvSAClassification":0.6102566666666668,"SynPerChatbotConvSAToneChatbotClassification":0.505537,"SynPerChatbotConvSAToneUserClassification":0.488459,"SynPerChatbotRAGFAQPC":0.644289,"SynPerChatbotRAGFAQRetrieval":0.32038,"SynPerChatbotRAGSumSRetrieval":0.532059,"SynPerChatbotSatisfactionLevelClassification":0.24717,"SynPerChatbotSumSRetrieval":0.321293,"SynPerQAPC":0.937589,"SynPerQARetrieval":0.86267,"SynPerSTS":0.875929,"SynPerTextKeywordsPC":0.970438,"SynPerTextToneClassification.v3":0.726264,"TRECCOVID-Fa.v2":0.73381,"WebFAQRetrieval":0.77262,"WikipediaRerankingMultilingual":0.8820753968253968,"WikipediaRetrievalMultilingual":0.89322} scoresByTaskType {"BitextMining":0.6107016666666666,"Classification":0.6531267604166666,"Clustering":0.5773111999999999,"PairClassification":0.8322472857142856,"Reranking":0.7456376984126984,"Retrieval":0.5892258823529412,"STS":0.796694} trainedOnTasks ["HotpotQA-FaHardNegatives","MIRACLReranking","MIRACLRetrievalHardNegatives","MSMARCO-FaHardNegatives","NQ-FaHardNegatives"] ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(fas%2C%20v2)/scores) Source MTEB Original model label BAAI/bge-m3 Observed 2026-10-21T16:40:30.698Z Evidence SHA-256 `37f3202cf21bcdda5c3dcff6173d2b935419ec840adc6e6b32d669920b744d43` Source locator `/rows/7` |  |  |  |
| 2 | [Multilingual-E5-Large](https://benchize.com/models/multilingual-e5-large.md) | 0.64source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.6427799123931626 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.6735584020524877 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 88 ### Reported statistics experiments Not reported rank 11 scoresByTask {"ArguAna-Fa.v2":0.41271,"BeytooteClustering":0.615017,"DeepSentiPers.v2":0.576946,"DigikalamagClassification":0.87054,"DigikalamagClustering":0.39889,"FEVER-FaHardNegatives":0.46152,"FarsTail":0.72554,"FarsiParaphraseDetection":0.975728,"Farsick":0.706725,"FiQA2018-Fa.v2":0.29464,"HamshahriClustring":0.674206,"HotpotQA-FaHardNegatives":0.61532,"MIRACLReranking":0.59361,"MIRACLRetrievalHardNegatives":0.59229,"MSMARCO-FaHardNegatives":0.68705,"MassiveIntentClassification":0.654879,"MassiveScenarioClassification":0.6858690000000001,"NLPTwitterAnalysisClassification.v2":0.765865,"NLPTwitterAnalysisClustering":0.784829,"NQ-FaHardNegatives":0.49834,"NeuCLIR2023RetrievalHardNegatives":0.50592,"ParsinluEntail":0.654551,"ParsinluQueryParaphPC":0.878273,"PerShopDomainClassification":0.551665,"PerShopIntentClassification":0.906917,"PersianFoodSentimentClassification":0.821167,"PersianTextEmotion.v2":0.609051,"PersianWebDocumentRetrieval":0.46761,"QuoraRetrieval-Fa.v2":0.77876,"SAMSumFa":0.924173,"SCIDOCS-Fa.v2":0.12217,"SIDClassification.v2":0.613742,"SIDClustring":0.386513,"SciFact-Fa.v2":0.60374,"StyleClassification":0.649219,"SynPerChatbotConvSAClassification":0.6076754444444444,"SynPerChatbotConvSAToneChatbotClassification":0.58072,"SynPerChatbotConvSAToneUserClassification":0.525951,"SynPerChatbotRAGFAQPC":0.630306,"SynPerChatbotRAGFAQRetrieval":0.23481,"SynPerChatbotRAGSumSRetrieval":0.498072,"SynPerChatbotSatisfactionLevelClassification":0.25231,"SynPerChatbotSumSRetrieval":0.275999,"SynPerQAPC":0.951645,"SynPerQARetrieval":0.87354,"SynPerSTS":0.879773,"SynPerTextKeywordsPC":0.94786,"SynPerTextToneClassification.v3":0.841189,"TRECCOVID-Fa.v2":0.71766,"WebFAQRetrieval":0.74585,"WikipediaRerankingMultilingual":0.8932,"WikipediaRetrievalMultilingual":0.90401} scoresByTaskType {"BitextMining":0.5660813333333333,"Classification":0.6571065902777777,"Clustering":0.5718909999999999,"PairClassification":0.8234147142857143,"Reranking":0.743405,"Retrieval":0.5597611764705883,"STS":0.7932490000000001} trainedOnTasks ["FEVER-FaHardNegatives","HotpotQA-FaHardNegatives","MIRACLReranking","MIRACLRetrievalHardNegatives","MSMARCO-FaHardNegatives","NQ-FaHardNegatives"] ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(fas%2C%20v2)/scores) Source MTEB Original model label intfloat/multilingual-e5-large Observed 2026-10-21T16:40:30.698Z Evidence SHA-256 `37f3202cf21bcdda5c3dcff6173d2b935419ec840adc6e6b32d669920b744d43` Source locator `/rows/10` |  |  |  |
| 3 | [Text Embedding 3 Small](https://benchize.com/models/text-embedding-3-small.md) | 0.55source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.549018875 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.5895871515131053 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct -1 ### Reported statistics experiments Not reported rank 26 scoresByTask {"ArguAna-Fa.v2":0.33277,"BeytooteClustering":0.603813,"DeepSentiPers.v2":0.50438,"DigikalamagClassification":0.849296,"DigikalamagClustering":0.459708,"FEVER-FaHardNegatives":0.27218,"FarsTail":0.68851,"FarsiParaphraseDetection":0.948317,"Farsick":0.608489,"FiQA2018-Fa.v2":0.10107,"HamshahriClustring":0.663308,"HotpotQA-FaHardNegatives":0.2813,"MIRACLReranking":0.34765,"MIRACLRetrievalHardNegatives":0.27242,"MSMARCO-FaHardNegatives":0.47673,"MassiveIntentClassification":0.5217075,"MassiveScenarioClassification":0.566567,"NLPTwitterAnalysisClassification.v2":0.720301,"NLPTwitterAnalysisClustering":0.789225,"NQ-FaHardNegatives":0.19959,"NeuCLIR2023RetrievalHardNegatives":0.34517,"ParsinluEntail":0.574657,"ParsinluQueryParaphPC":0.770129,"PerShopDomainClassification":0.492143,"PerShopIntentClassification":0.88702,"PersianFoodSentimentClassification":0.6614990000000001,"PersianTextEmotion.v2":0.445909,"PersianWebDocumentRetrieval":0.35082,"QuoraRetrieval-Fa.v2":0.62235,"SAMSumFa":0.846083,"SCIDOCS-Fa.v2":0.08402,"SIDClassification.v2":0.527163,"SIDClustring":0.35678,"SciFact-Fa.v2":0.41255,"StyleClassification":0.622917,"SynPerChatbotConvSAClassification":0.715087,"SynPerChatbotConvSAToneChatbotClassification":0.666578,"SynPerChatbotConvSAToneUserClassification":0.603736,"SynPerChatbotRAGFAQPC":0.689606,"SynPerChatbotRAGFAQRetrieval":0.2826,"SynPerChatbotRAGSumSRetrieval":0.478077,"SynPerChatbotSatisfactionLevelClassification":0.366428,"SynPerChatbotSumSRetrieval":0.221603,"SynPerQAPC":0.905553,"SynPerQARetrieval":0.63577,"SynPerSTS":0.773342,"SynPerTextKeywordsPC":0.914576,"SynPerTextToneClassification.v3":0.728859,"TRECCOVID-Fa.v2":0.29366,"WebFAQRetrieval":0.5044,"WikipediaRerankingMultilingual":0.809355,"WikipediaRetrievalMultilingual":0.75321} scoresByTaskType {"BitextMining":0.5152543333333334,"Classification":0.6174744062499999,"Clustering":0.5745668,"PairClassification":0.7844782857142858,"Reranking":0.5785025,"Retrieval":0.36591823529411766,"STS":0.6909155} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/MTEB(fas%2C%20v2)/scores) Source MTEB Original model label openai/text-embedding-3-small Observed 2026-10-21T16:40:30.698Z Evidence SHA-256 `37f3202cf21bcdda5c3dcff6173d2b935419ec840adc6e6b32d669920b744d43` Source locator `/rows/25` |  |  |  |

---
Cite as: MTEB(fas, v2) · Persian — Benchize, https://benchize.com/benchmarks/mteb%3AMTEB(fas%2C%20v2) (retrieved 2026-10-21)