multimodal benchmark
MIEB(eng) Multimodal Leaderboard
Direct model-level leaderboard results from MTEB.
Updated Aug 18, 2026
Models83
Model coverage83
MetricOfficial leaderboard task-type mean
EvidenceA
MIEB(eng) Multimodal Ranking
Higher official leaderboard task-type mean ranks better on this benchmark.
Sort by
Rank | Model | Official leaderboard task-type mean | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelDUArgus-Colqwen3.5-9b-v0Datascience Uibk | Official leaderboard task-type mean92.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank02 | ModelNV | Official leaderboard task-type mean92.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank03 | ModelDUArgus-Colqwen3.5-9b-v0-bf16Datascience Uibk | Official leaderboard task-type mean92.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank04 | ModelDUArgus-Colqwen3.5-4b-v0Datascience Uibk | Official leaderboard task-type mean92.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank05 | ModelVUVultronRetrieverCore-Qwen3.5-4.5BVultr | Official leaderboard task-type mean92.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank06 | ModelDUArgus-Colqwen3.5-4b-v0-bf16Datascience Uibk | Official leaderboard task-type mean92.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank07 | ModelVUVultronRetrieverPrime-Qwen3.5-8BVultr | Official leaderboard task-type mean92.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank08 | ModelNV | Official leaderboard task-type mean91.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank09 | ModelNV | Official leaderboard task-type mean91.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank10 | ModelAScolqwen3.5-4.5B-v3Athrael Soju | Official leaderboard task-type mean91.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank11 | ModelDUArgus-Colqwen3.5-2b-v0Datascience Uibk | Official leaderboard task-type mean91.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank12 | ModelDUArgus-Colqwen3.5-2b-v0-bf16Datascience Uibk | Official leaderboard task-type mean91.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank13 | ModelOAOps-Colqwen3-4BOpensearch Ai | Official leaderboard task-type mean91.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank14 | ModelWOwebAI-ColVec1-9bWebai Official | Official leaderboard task-type mean91.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank15 | ModelVASauerkrautLM-ColQwen3-8b-v0.1Vagosolutions | Official leaderboard task-type mean91.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank16 | ModelNV | Official leaderboard task-type mean91.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank17 | ModelVASauerkrautLM-ColQwen3-4b-v0.1Vagosolutions | Official leaderboard task-type mean90.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank18 | ModelTOtomoro-colqwen3-embed-8bTomoroai | Official leaderboard task-type mean90.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank19 | ModelAPEvoQwen2.5-VL-Retriever-7B-v1Apsarastackmaas | Official leaderboard task-type mean90.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank20 | ModelAPEvoQwen2.5-VL-Retriever-3B-v1Apsarastackmaas | Official leaderboard task-type mean90.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank21 | ModelTOtomoro-colqwen3-embed-4bTomoroai | Official leaderboard task-type mean90.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank22 | ModelNV | Official leaderboard task-type mean90.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank23 | ModelWOwebAI-ColVec1-4bWebai Official | Official leaderboard task-type mean90.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank24 | ModelJIjina-embeddings-v4Jinaai | Official leaderboard task-type mean90.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank25 | ModelVASauerkrautLM-ColQwen3-2b-v0.1Vagosolutions | Official leaderboard task-type mean90.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank26 | ModelNAcolnomic-embed-multimodal-3bNomic Ai | Official leaderboard task-type mean89.9% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank27 | ModelNAcolnomic-embed-multimodal-7bNomic Ai | Official leaderboard task-type mean89.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank28 | ModelVIcolqwen2.5-v0.2Vidore | Official leaderboard task-type mean89.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank29 | ModelVIcolqwen2-v1.0Vidore | Official leaderboard task-type mean89.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank30 | ModelVASauerkrautLM-ColQwen3-1.7b-Turbo-v0.1Vagosolutions | Official leaderboard task-type mean88.9% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank31 | ModelVUVultronRetrieverFlash-Qwen3.5-0.8BVultr | Official leaderboard task-type mean88.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank32 | ModelIGgranite-vision-3.3-2b-embeddingIbm Granite | Official leaderboard task-type mean86.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank33 | ModelEAeager-embed-v1Eagerworks | Official leaderboard task-type mean84.9% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank34 | ModelVIcolpali-v1.3Vidore | Official leaderboard task-type mean84.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank35 | ModelVASauerkrautLM-ColLFM2-450M-v0.1Vagosolutions | Official leaderboard task-type mean83.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank36 | ModelVIcolpali-v1.2Vidore | Official leaderboard task-type mean83.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank37 | ModelVIcolSmol-500MVidore | Official leaderboard task-type mean82.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank38 | ModelNANanoVDR-S-MultiNanovdr | Official leaderboard task-type mean82.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank39 | ModelVASauerkrautLM-ColMinistral3-3b-v0.1Vagosolutions | Official leaderboard task-type mean82.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank40 | ModelVIcolpali-v1.1Vidore | Official leaderboard task-type mean81.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank41 | ModelVIcolSmol-256MVidore | Official leaderboard task-type mean79.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank42 | ModelVOvoyage-multimodal-3Voyageai | Official leaderboard task-type mean61.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank43 | ModelJIjina-embeddings-v5-omni-smallJinaai | Official leaderboard task-type mean58.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank44 | ModelJIjina-embeddings-v5-omni-nanoJinaai | Official leaderboard task-type mean55.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank45 | ModelGO | Official leaderboard task-type mean48.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank46 | ModelROe5-vRoyokong | Official leaderboard task-type mean48.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank47 | ModelGO | Official leaderboard task-type mean46.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank48 | ModelGO | Official leaderboard task-type mean45.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank49 | ModelLACLIP-ViT-bigG-14-laion2B-39B-b160kLaion | Official leaderboard task-type mean42.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank50 | ModelGO | Official leaderboard task-type mean41.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank51 | ModelLACLIP-ViT-H-14-laion2B-s32B-b79KLaion | Official leaderboard task-type mean40.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank52 | ModelGO | Official leaderboard task-type mean39.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank53 | ModelLACLIP-ViT-g-14-laion2B-s34B-b88KLaion | Official leaderboard task-type mean38.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank54 | ModelTLVLM2Vec-FullTiger Lab | Official leaderboard task-type mean38.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank55 | ModelTLVLM2Vec-LoRATiger Lab | Official leaderboard task-type mean38.7% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank56 | ModelLACLIP-ViT-L-14-DataComp.XL-s13B-b90KLaion | Official leaderboard task-type mean38.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank57 | ModelLACLIP-ViT-L-14-laion2B-s32B-b82KLaion | Official leaderboard task-type mean37.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank58 | ModelQUEVA02-CLIP-bigE-14-plusQuansun | Official leaderboard task-type mean36.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank59 | ModelOP | Official leaderboard task-type mean35.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank60 | ModelQUEVA02-CLIP-bigE-14Quansun | Official leaderboard task-type mean35.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank61 | ModelGO | Official leaderboard task-type mean34.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank62 | ModelGO | Official leaderboard task-type mean31.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank63 | ModelKAalign-baseKakaobrain | Official leaderboard task-type mean31.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank64 | ModelGO | Official leaderboard task-type mean31.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank65 | ModelLACLIP-ViT-B-16-DataComp.XL-s13B-b90KLaion | Official leaderboard task-type mean30.9% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank66 | ModelQUEVA02-CLIP-L-14Quansun | Official leaderboard task-type mean28.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank67 | ModelOP | Official leaderboard task-type mean27.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank68 | ModelSAblip-itm-large-cocoSalesforce | Official leaderboard task-type mean25.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank69 | ModelSAblip-itm-large-flickrSalesforce | Official leaderboard task-type mean25.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank70 | ModelLACLIP-ViT-B-32-laion2B-s34B-b79KLaion | Official leaderboard task-type mean25.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank71 | ModelJIjina-clip-v1Jinaai | Official leaderboard task-type mean25.0% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank72 | ModelSAblip-itm-base-cocoSalesforce | Official leaderboard task-type mean24.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank73 | ModelLACLIP-ViT-B-32-DataComp.XL-s13B-b90KLaion | Official leaderboard task-type mean22.3% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank74 | ModelSAblip-itm-base-flickrSalesforce | Official leaderboard task-type mean21.6% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank75 | ModelNAnomic-embed-vision-v1.5Nomic Ai | Official leaderboard task-type mean21.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank76 | ModelBAbge-visualized-m3Baai | Official leaderboard task-type mean20.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank77 | ModelOP | Official leaderboard task-type mean20.4% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank78 | ModelQUEVA02-CLIP-B-16Quansun | Official leaderboard task-type mean20.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank79 | ModelSAblip2-opt-6.7b-cocoSalesforce | Official leaderboard task-type mean20.1% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank80 | ModelSAblip2-opt-2.7bSalesforce | Official leaderboard task-type mean19.5% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank81 | ModelBAbge-visualized-baseBaai | Official leaderboard task-type mean19.2% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank82 | ModelSAblip-image-captioning-baseSalesforce | Official leaderboard task-type mean0.9% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
| Rank83 | ModelSAblip-image-captioning-largeSalesforce | Official leaderboard task-type mean0.8% | PercentileN/A | Participants83 | EvidenceA | Evaluated |
MIEB(eng) Multimodal Highlights
The leading models and scores on this benchmark.
MIEB(eng) Multimodal Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MIEB(eng) Multimodal
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis mieb(eng) multimodal 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.
- 01DUArgus-Colqwen3.5-9b-v0Datascience Uibk
- Official leaderboard task-type mean
- 92.7%
Strengths
- Ranks #1 of 83 compared models
- A evidence result
Considerations
- This result measures MIEB(eng) Multimodal, not total model capability
- 02NV
- Official leaderboard task-type mean
- 92.7%
Strengths
- Ranks #2 of 83 compared models
- A evidence result
Considerations
- This result measures MIEB(eng) Multimodal, not total model capability
- 03DUArgus-Colqwen3.5-9b-v0-bf16Datascience Uibk
- Official leaderboard task-type mean
- 92.6%
Strengths
- Ranks #3 of 83 compared models
- A evidence result
Considerations
- This result measures MIEB(eng) Multimodal, not total model capability
- 04DUArgus-Colqwen3.5-4b-v0Datascience Uibk
- Official leaderboard task-type mean
- 92.3%
Strengths
- Ranks #4 of 83 compared models
- A evidence result
Considerations
- This result measures MIEB(eng) Multimodal, not total model capability
- 05
- Official leaderboard task-type mean
- 92.2%
Strengths
- Ranks #5 of 83 compared models
- A evidence result
Considerations
- This result measures MIEB(eng) Multimodal, not total model capability
Selection summary
Best AI Models for MIEB(eng) Multimodal
Argus-Colqwen3.5-9b-v0 currently leads MIEB(eng) Multimodal with 92.7%. 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.
What is MIEB(eng) Multimodal?
What MIEB(eng) Multimodal 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
- MIEB(eng):multimodal
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- mteb-mieb-eng-multimodal|mteb|current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MIEB(eng) Multimodal.
Which model scores highest on MIEB(eng) Multimodal?
Argus-Colqwen3.5-9b-v0 is currently ranked first with 92.7%.
What does MIEB(eng) Multimodal 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?
83 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.