multimodal benchmark
MIEB English Multimodal is a multimodal benchmark listed on the current MIEB(eng) leaderboard.
Updated Sep 8, 2026
Higher official leaderboard task-type mean ranks better on this benchmark.
Rank | Model | Official leaderboard task-type mean | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelDUArgus-Colqwen3.5-9b-v0Datascience Uibk | Official leaderboard task-type mean92.67% | Percentile100.00% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank02 | ModelNV | Official leaderboard task-type mean92.65% | Percentile98.80% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank03 | ModelDUArgus-Colqwen3.5-9b-v0-bf16Datascience Uibk | Official leaderboard task-type mean92.59% | Percentile97.59% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank04 | ModelDUArgus-Colqwen3.5-4b-v0Datascience Uibk | Official leaderboard task-type mean92.30% | Percentile96.39% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank05 | ModelVUVultronRetrieverCore-Qwen3.5-4.5BVultr | Official leaderboard task-type mean92.21% | Percentile95.18% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank06 | ModelDUArgus-Colqwen3.5-4b-v0-bf16Datascience Uibk | Official leaderboard task-type mean92.14% | Percentile93.98% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank07 | ModelVUVultronRetrieverPrime-Qwen3.5-8BVultr | Official leaderboard task-type mean92.08% | Percentile92.77% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank08 | ModelNV | Official leaderboard task-type mean91.74% | Percentile91.57% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank09 | ModelNV | Official leaderboard task-type mean91.62% | Percentile90.36% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank10 | ModelAScolqwen3.5-4.5B-v3Athrael Soju | Official leaderboard task-type mean91.54% | Percentile89.16% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank11 | ModelDUArgus-Colqwen3.5-2b-v0Datascience Uibk | Official leaderboard task-type mean91.49% | Percentile87.95% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank12 | ModelDUArgus-Colqwen3.5-2b-v0-bf16Datascience Uibk | Official leaderboard task-type mean91.49% | Percentile86.75% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank13 | ModelOAOps-Colqwen3-4BOpensearch Ai | Official leaderboard task-type mean91.36% | Percentile85.54% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank14 | ModelWOwebAI-ColVec1-9bWebai Official | Official leaderboard task-type mean91.30% | Percentile84.34% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank15 | ModelVASauerkrautLM-ColQwen3-8b-v0.1Vagosolutions | Official leaderboard task-type mean91.08% | Percentile83.13% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank16 | ModelNV | Official leaderboard task-type mean91.00% | Percentile81.93% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank17 | ModelVASauerkrautLM-ColQwen3-4b-v0.1Vagosolutions | Official leaderboard task-type mean90.80% | Percentile80.72% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank18 | ModelTOtomoro-colqwen3-embed-8bTomoroai | Official leaderboard task-type mean90.76% | Percentile79.52% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank19 | ModelAPEvoQwen2.5-VL-Retriever-7B-v1Apsarastackmaas | Official leaderboard task-type mean90.68% | Percentile78.31% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank20 | ModelAPEvoQwen2.5-VL-Retriever-3B-v1Apsarastackmaas | Official leaderboard task-type mean90.67% | Percentile77.11% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank21 | ModelTOtomoro-colqwen3-embed-4bTomoroai | Official leaderboard task-type mean90.57% | Percentile75.90% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank22 | ModelNV | Official leaderboard task-type mean90.50% | Percentile74.70% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank23 | ModelWOwebAI-ColVec1-4bWebai Official | Official leaderboard task-type mean90.49% | Percentile73.49% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank24 | ModelJIjina-embeddings-v4Jinaai | Official leaderboard task-type mean90.35% | Percentile72.29% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank25 | ModelVASauerkrautLM-ColQwen3-2b-v0.1Vagosolutions | Official leaderboard task-type mean90.24% | Percentile71.08% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank26 | ModelNAcolnomic-embed-multimodal-3bNomic Ai | Official leaderboard task-type mean89.86% | Percentile69.88% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank27 | ModelNAcolnomic-embed-multimodal-7bNomic Ai | Official leaderboard task-type mean89.72% | Percentile68.67% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank28 | ModelVIcolqwen2.5-v0.2Vidore | Official leaderboard task-type mean89.54% | Percentile67.47% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank29 | ModelVIcolqwen2-v1.0Vidore | Official leaderboard task-type mean89.23% | Percentile66.27% | Participants84 | EvidenceA | EvaluatedN/A |
| Rank30 | ModelVASauerkrautLM-ColQwen3-1.7b-Turbo-v0.1Vagosolutions | Official leaderboard task-type mean88.89% | Percentile65.06% | Participants84 | EvidenceA | EvaluatedN/A |
The leading models and scores on this benchmark.
A closer view of the leading scores on this benchmark.
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What MIEB English Multimodal measures and how its scores work.
MIEB English Multimodal is a multimodal benchmark.
It reports the Official leaderboard task-type mean as a ratio.
Scores are shown in ratio. This benchmark is verified and has an evidence level of A.
MTEB. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
Common questions about MIEB English Multimodal.
Argus-Colqwen3.5-9b-v0 is currently ranked first with 92.67%.
The current leaders are Argus-Colqwen3.5-9b-v0 (92.67%), nemotron-colembed-vl-8b-v2 (92.65%), and Argus-Colqwen3.5-9b-v0-bf16 (92.59%).
No matched official input price is currently available.
No matched runtime record is currently available.
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.
It reports the Official leaderboard task-type mean as a ratio.
Yes. Higher values rank better for this benchmark.
84 model results are currently shown.
No. This benchmark is shown for reference but does not contribute to the overall score.
Ranking basisThis mieb english multimodal AI model leaderboard uses descending official leaderboard task-type mean in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
Argus-Colqwen3.5-9b-v0 currently leads MIEB English Multimodal with 92.67%. 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.