llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

retrieval benchmark

MTEB(eng, v2) Retrieval Leaderboard

Direct model-level leaderboard results from MTEB.

Updated Aug 18, 2026

Models100
Model coverage208
MetricOfficial leaderboard task-type mean
EvidenceA

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

MTEB(eng, v2) Retrieval Ranking

Higher official leaderboard task-type mean ranks better on this benchmark.

30 of 100 rows
Columns

Show columns

Sort by
Rank
Model
Official leaderboard task-type mean
Percentile
Participants
Evidence
Evaluated
Rank01ModelIEYuan-embedding-2.0-enIeityuanOfficial leaderboard task-type mean70.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank02ModelJCingot-8b-r3JcornersOfficial leaderboard task-type mean70.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank03ModelACQwen3-Embedding-8BAlibaba Cloud / Qwen TeamOfficial leaderboard task-type mean69.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank04ModelGAgeevec-embeddings-1.0Geevec AiOfficial leaderboard task-type mean68.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank05ModelACQwen3-Embedding-4BAlibaba Cloud / Qwen TeamOfficial leaderboard task-type mean68.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank06ModelSESeed1.5-EmbeddingOfficial leaderboard task-type mean67.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank07ModelKLQZhou-EmbeddingKingsoft LlmOfficial leaderboard task-type mean67.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank08ModelINJasper-Token-Compression-600MInfgradOfficial leaderboard task-type mean66.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank09ModelANLGAI-Embedding-PreviewAnnamodelsOfficial leaderboard task-type mean66.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank10ModelBYSeed1.6-embeddingByteDanceOfficial leaderboard task-type mean64.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank11ModelGOGemini Embedding 001GoogleOfficial leaderboard task-type mean64.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank12ModelINinf-retriever-v1InflyOfficial leaderboard task-type mean64.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank13ModelNVNV-Embed-v2NVIDIAOfficial leaderboard task-type mean62.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank14ModelGAgeevec-embeddings-1.0-liteGeevec AiOfficial leaderboard task-type mean62.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank15ModelACQwen3-Embedding-0.6BAlibaba Cloud / Qwen TeamOfficial leaderboard task-type mean61.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank16ModelGRGeoEmbeddingGeogpt Research ProjectOfficial leaderboard task-type mean60.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank17ModelINinf-retriever-v1-1.5bInflyOfficial leaderboard task-type mean60.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank18ModelCAF2LLM-v2-14BCodefuse AiOfficial leaderboard task-type mean60.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank19ModelLALinq-Embed-MistralLinq Ai ResearchOfficial leaderboard task-type mean60.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank20ModelNVNV-Embed-v1NVIDIAOfficial leaderboard task-type mean60.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank21ModelJIjina-embeddings-v5-omni-smallJinaaiOfficial leaderboard task-type mean60.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank22ModelJIjina-embeddings-v5-text-smallJinaaiOfficial leaderboard task-type mean60.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank23ModelCAF2LLM-v2-8BCodefuse AiOfficial leaderboard task-type mean59.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank24ModelASGiga-Embeddings-instructAi SageOfficial leaderboard task-type mean59.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank25ModelCAF2LLM-4BCodefuse AiOfficial leaderboard task-type mean59.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank26ModelITICT-TIME-and-Querit-embedding-v1Ict Time And QueritOfficial leaderboard task-type mean59.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank27ModelSASFR-Embedding-MistralSalesforceOfficial leaderboard task-type mean59.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank28ModelCAF2LLM-v2-4BCodefuse AiOfficial leaderboard task-type mean59.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank29ModelGOtext-embedding-004GoogleOfficial leaderboard task-type mean59.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank30ModelSNsnowflake-arctic-embed-lSnowflakeOfficial leaderboard task-type mean59.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank31ModelJIjina-embeddings-v5-omni-nanoJinaaiOfficial leaderboard task-type mean58.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank32ModelJIjina-embeddings-v5-text-nanoJinaaiOfficial leaderboard task-type mean58.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank33ModelGOtext-embedding-005GoogleOfficial leaderboard task-type mean58.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank34ModelSNsnowflake-arctic-embed-l-v2.0SnowflakeOfficial leaderboard task-type mean58.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank35ModelKEKaLM-embedding-multilingual-mini-instruct-v2.5Kalm EmbeddingOfficial leaderboard task-type mean58.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank36ModelSNsnowflake-arctic-embed-m-v2.0SnowflakeOfficial leaderboard task-type mean58.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank37ModelANgte-Qwen2-7B-instructAlibaba NlpOfficial leaderboard task-type mean58.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank38ModelSNsnowflake-arctic-embed-m-v1.5SnowflakeOfficial leaderboard task-type mean58.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank39ModelOPtext-embedding-3-largeOpenAIOfficial leaderboard task-type mean58.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank40ModelCAF2LLM-1.7BCodefuse AiOfficial leaderboard task-type mean57.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank41ModelINe5-mistral-7b-instructIntfloatOfficial leaderboard task-type mean57.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank42ModelSNsnowflake-arctic-embed-mSnowflakeOfficial leaderboard task-type mean57.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank43ModelSNsnowflake-arctic-embed-m-longSnowflakeOfficial leaderboard task-type mean57.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank44ModelITBOOM_4B_v1Ict Time And QueritOfficial leaderboard task-type mean57.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank45ModelCAF2LLM-v2-1.7BCodefuse AiOfficial leaderboard task-type mean56.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank46ModelIGgranite-embedding-english-r2Ibm GraniteOfficial leaderboard task-type mean56.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank47ModelANgte-Qwen1.5-7B-instructAlibaba NlpOfficial leaderboard task-type mean56.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank48ModelJIjina-embeddings-v4JinaaiOfficial leaderboard task-type mean56.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank49ModelNOjasper_en_vision_language_v1NovasearchOfficial leaderboard task-type mean56.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank50ModelWHUAE-Large-V1WhereisaiOfficial leaderboard task-type mean55.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank51ModelCAF2LLM-0.6BCodefuse AiOfficial leaderboard task-type mean55.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank52ModelGOembeddinggemma-300mGoogleOfficial leaderboard task-type mean55.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank53ModelCOCohere-embed-multilingual-v3.0CohereOfficial leaderboard task-type mean55.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank54ModelCOCohere-embed-english-v3.0CohereOfficial leaderboard task-type mean55.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank55ModelIGgranite-embedding-125m-englishIbm GraniteOfficial leaderboard task-type mean55.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank56ModelBAbge-large-en-v1.5BaaiOfficial leaderboard task-type mean55.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank57ModelMAmxbai-embed-large-v1Mixedbread AiOfficial leaderboard task-type mean55.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank58ModelTATarka-Embedding-350M-V1Tarka AirOfficial leaderboard task-type mean55.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank59ModelBIBidirLM-Omni-2.5B-EmbeddingBidirlmOfficial leaderboard task-type mean55.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank60ModelGRGritLM-7BGritlmOfficial leaderboard task-type mean54.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank61ModelSNsnowflake-arctic-embed-sSnowflakeOfficial leaderboard task-type mean54.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank62ModelNAnomic-embed-text-v2-moeNomic AiOfficial leaderboard task-type mean54.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank63ModelBAbge-base-en-v1.5BaaiOfficial leaderboard task-type mean54.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank64ModelGOtext-multilingual-embedding-002GoogleOfficial leaderboard task-type mean54.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank65ModelAVGIST-large-Embedding-v0AvsolatorioOfficial leaderboard task-type mean54.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank66ModelCAF2LLM-v2-0.6BCodefuse AiOfficial leaderboard task-type mean54.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank67ModelJIjina-embeddings-v3JinaaiOfficial leaderboard task-type mean54.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank68ModelHTKaLM-embedding-multilingual-mini-v1Hit TmgOfficial leaderboard task-type mean54.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank69ModelIGgranite-embedding-small-english-r2Ibm GraniteOfficial leaderboard task-type mean53.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank70ModelBAbge-small-en-v1.5BaaiOfficial leaderboard task-type mean53.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank71ModelSASFR-Embedding-2_RSalesforceOfficial leaderboard task-type mean53.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank72ModelAVNoInstruct-small-Embedding-v0AvsolatorioOfficial leaderboard task-type mean53.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank73ModelAVGIST-Embedding-v0AvsolatorioOfficial leaderboard task-type mean53.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank74ModelOPtext-embedding-3-smallOpenAIOfficial leaderboard task-type mean53.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank75ModelINmultilingual-e5-large-instructIntfloatOfficial leaderboard task-type mean53.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank76ModelVOvoyage-3VoyageaiOfficial leaderboard task-type mean53.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank77ModelCAF2LLM-v2-330MCodefuse AiOfficial leaderboard task-type mean53.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank78ModelTHgte-largeThenlperOfficial leaderboard task-type mean53.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank79ModelPRmini-gtePrdevOfficial leaderboard task-type mean53.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank80ModelCOCohere-embed-english-light-v3.0CohereOfficial leaderboard task-type mean53.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank81ModelABMedEmbed-small-v0.1AbhinandOfficial leaderboard task-type mean52.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank82ModelNOstella_en_400M_v5NovasearchOfficial leaderboard task-type mean52.7%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank83ModelSNsnowflake-arctic-embed-xsSnowflakeOfficial leaderboard task-type mean52.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank84ModelIGgranite-embedding-311m-multilingual-r2Ibm GraniteOfficial leaderboard task-type mean52.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank85ModelGRGritLM-8x7BGritlmOfficial leaderboard task-type mean52.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank86ModelNOstella_en_1.5B_v5NovasearchOfficial leaderboard task-type mean52.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank87ModelVPSearchMap_PreviewVplabsOfficial leaderboard task-type mean52.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank88ModelTEOmniEmbed-v0.1TevatronOfficial leaderboard task-type mean52.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank89ModelCOCohere-embed-multilingual-light-v3.0CohereOfficial leaderboard task-type mean52.1%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank90ModelAVGIST-small-Embedding-v0AvsolatorioOfficial leaderboard task-type mean52.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank91ModelMAdeepset-mxbai-embed-de-large-v1Mixedbread AiOfficial leaderboard task-type mean52.0%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank92ModelTHgte-baseThenlperOfficial leaderboard task-type mean51.9%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank93ModelIGgranite-embedding-30m-englishIbm GraniteOfficial leaderboard task-type mean51.8%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank94ModelINstella-base-en-v2InfgradOfficial leaderboard task-type mean51.6%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank95ModelINmultilingual-e5-largeIntfloatOfficial leaderboard task-type mean51.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank96ModelTATarka-Embedding-150M-V1Tarka AirOfficial leaderboard task-type mean51.5%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank97ModelIGgranite-embedding-278m-multilingualIbm GraniteOfficial leaderboard task-type mean51.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank98ModelMOmdbr-leaf-mtMongodbOfficial leaderboard task-type mean51.4%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank99ModelMNLLM2Vec-Mistral-7B-Instruct-v2-mntp-supervisedMcgill NlpOfficial leaderboard task-type mean51.3%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026
Rank100ModelHCe5-omni-3BHaon ChenOfficial leaderboard task-type mean51.2%PercentileN/AParticipants208EvidenceAEvaluatedAug 18, 2026

MTEB(eng, v2) Retrieval Highlights

The leading models and scores on this benchmark.

Rank #1Yuan-embedding-2.0-en70.7%Rank #2ingot-8b-r370.0%Rank #3Qwen3-Embedding-8B69.4%Rank #4geevec-embeddings-1.068.8%

MTEB(eng, v2) Retrieval Score Distribution

A closer view of the leading scores on this benchmark.

MTEB(eng, v2) Retrieval

The Top AI Models for MTEB(eng, v2) Retrieval

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

Ranking basisThis mteb(eng, v2) retrieval AI model leaderboard uses descending official leaderboard task-type mean in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    IE
    Yuan-embedding-2.0-enIeityuan
    Official leaderboard task-type mean
    70.7%

    Strengths

    • Ranks #1 of 208 compared models
    • A evidence result

    Considerations

    • This result measures MTEB(eng, v2) Retrieval, not total model capability
  2. 02
    JC
    ingot-8b-r3Jcorners
    Official leaderboard task-type mean
    70.0%

    Strengths

    • Ranks #2 of 208 compared models
    • A evidence result

    Considerations

    • This result measures MTEB(eng, v2) Retrieval, not total model capability
  3. 03
    AC
    Qwen3-Embedding-8BAlibaba Cloud / Qwen Team
    Official leaderboard task-type mean
    69.4%

    Strengths

    • Ranks #3 of 208 compared models
    • A evidence result

    Considerations

    • This result measures MTEB(eng, v2) Retrieval, not total model capability
  4. 04
    GA
    geevec-embeddings-1.0Geevec Ai
    Official leaderboard task-type mean
    68.8%

    Strengths

    • Ranks #4 of 208 compared models
    • A evidence result

    Considerations

    • This result measures MTEB(eng, v2) Retrieval, not total model capability
  5. 05
    AC
    Qwen3-Embedding-4BAlibaba Cloud / Qwen Team
    Official leaderboard task-type mean
    68.5%

    Strengths

    • Ranks #5 of 208 compared models
    • A evidence result

    Considerations

    • This result measures MTEB(eng, v2) Retrieval, not total model capability

Selection summary

Best AI Models for MTEB(eng, v2) Retrieval

Yuan-embedding-2.0-en currently leads MTEB(eng, v2) Retrieval with 70.7%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.

Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.

Benchmark rank #1Yuan-embedding-2.0-en70.7%Benchmark rank #2ingot-8b-r370.0%Benchmark rank #3Qwen3-Embedding-8B69.4%

What is MTEB(eng, v2) Retrieval?

What MTEB(eng, v2) Retrieval measures and how its scores work.

Direct model-level leaderboard results from MTEB.

Scores are shown in ratio. This benchmark is verified and has an evidence level of A.

Family
MTEB(eng, v2):retrieval
Modality
multimodal
Primary category
retrieval
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
mteb-mteb-eng-v2-retrieval|mteb|current

Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about MTEB(eng, v2) Retrieval.

Which model scores highest on MTEB(eng, v2) Retrieval?

Yuan-embedding-2.0-en is currently ranked first with 70.7%.

What does MTEB(eng, v2) Retrieval measure?

Direct model-level leaderboard results from MTEB.

Is a higher official leaderboard task-type mean better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

100 model results are currently shown.

Does this benchmark affect the overall score?

Yes. This benchmark can contribute to the current LLMBoard capability score.