retrieval benchmark
ViDoRe v3 Overall is the current overall leaderboard for ViDoRe(v3).
Updated Sep 8, 2026
Higher mean task score ranks better on this benchmark.
Rank | Model | Mean task score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelWOwebAI-ColVec1.1-8bWebai Official | Mean task score64.95% | Percentile100.00% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank02 | ModelVUVultronRetrieverPrime-Qwen3.5-8BVultr | Mean task score64.26% | Percentile97.44% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank03 | ModelWOwebAI-ColVec1.1-4bWebai Official | Mean task score63.90% | Percentile94.87% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank04 | ModelVUVultronRetrieverCore-Qwen3.5-4.5BVultr | Mean task score63.57% | Percentile92.31% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank05 | ModelNV | Mean task score63.42% | Percentile89.74% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank06 | ModelWOwebAI-ColVec1-9bWebai Official | Mean task score63.00% | Percentile87.18% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank07 | ModelWOwebAI-ColVec1-4bWebai Official | Mean task score62.22% | Percentile84.62% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank08 | ModelTOtomoro-colqwen3-embed-8bTomoroai | Mean task score61.59% | Percentile82.05% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank09 | ModelNV | Mean task score61.54% | Percentile79.49% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank10 | ModelAScolqwen3.5-4.5B-v3Athrael Soju | Mean task score61.46% | Percentile76.92% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank11 | ModelOAOps-Colqwen3-4BOpensearch Ai | Mean task score61.17% | Percentile74.36% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank12 | ModelTOtomoro-colqwen3-embed-4bTomoroai | Mean task score60.20% | Percentile71.79% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank13 | ModelNV | Mean task score59.79% | Percentile69.23% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank14 | ModelAC | Mean task score57.78% | Percentile66.67% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank15 | ModelNAcolnomic-embed-multimodal-7bNomic Ai | Mean task score57.33% | Percentile64.10% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank16 | ModelNV | Mean task score57.26% | Percentile61.54% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank17 | ModelVUVultronRetrieverFlash-Qwen3.5-0.8BVultr | Mean task score56.16% | Percentile58.97% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank18 | ModelNAcolnomic-embed-multimodal-3bNomic Ai | Mean task score55.78% | Percentile56.41% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank19 | ModelNV | Mean task score55.59% | Percentile53.85% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank20 | ModelAC | Mean task score52.64% | Percentile51.28% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank21 | ModelVIcolqwen2.5-v0.2Vidore | Mean task score51.90% | Percentile48.72% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank22 | ModelJIjina-embeddings-v4Jinaai | Mean task score48.52% | Percentile46.15% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank23 | ModelJIjina-embeddings-v5-omni-smallJinaai | Mean task score45.61% | Percentile43.59% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank24 | ModelJIjina-embeddings-v5-omni-nanoJinaai | Mean task score45.56% | Percentile41.03% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank25 | ModelVIcolqwen2-v1.0Vidore | Mean task score44.65% | Percentile38.46% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank26 | ModelVIcolpali-v1.3Vidore | Mean task score43.05% | Percentile35.90% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank27 | ModelMOcolmodernvbertModernvbert | Mean task score26.07% | Percentile33.33% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank28 | ModelVIcolSmol-256MVidore | Mean task score21.40% | Percentile30.77% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank29 | ModelGO | Mean task score16.44% | Percentile28.21% | Participants40 | EvidenceA | EvaluatedN/A |
| Rank30 | ModelGO | Mean task score15.75% | Percentile25.64% | Participants40 | 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 ViDoRe v3 Overall measures and how its scores work.
ViDoRe v3 Overall is a retrieval benchmark leaderboard.
It reports Mean task score 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 ViDoRe v3 Overall.
webAI-ColVec1.1-8b is currently ranked first with 64.95%.
The current leaders are webAI-ColVec1.1-8b (64.95%), VultronRetrieverPrime-Qwen3.5-8B (64.26%), and webAI-ColVec1.1-4b (63.90%).
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 Mean task score as a ratio.
Yes. Higher values rank better for this benchmark.
40 model results are currently shown.
No. This benchmark is shown for reference but does not contribute to the overall score.
Ranking basisThis vidore v3 overall AI model leaderboard uses descending mean task score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
webAI-ColVec1.1-8b currently leads ViDoRe v3 Overall with 64.95%. 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.