multimodal benchmark
DocVQAtest Leaderboard
DocVQAtest is a Visual Question Answering benchmark based on 12,000+ document images and 50,000 questions.
Updated Sep 5, 2026
Models11
Model coverage11
MetricScore
EvidenceB
DocVQAtest Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score97.10% | Percentile100.00% | Participants11 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score96.90% | Percentile90.00% | Participants11 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score96.50% | Percentile80.00% | Participants11 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score96.50% | Percentile70.00% | Participants11 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score96.10% | Percentile60.00% | Participants11 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score96.10% | Percentile50.00% | Participants11 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score95.30% | Percentile40.00% | Participants11 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score95.30% | Percentile30.00% | Participants11 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score95.00% | Percentile20.00% | Participants11 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score95.00% | Percentile10.00% | Participants11 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score94.20% | Percentile0.00% | Participants11 | EvidenceC | Evaluated |
DocVQAtest Highlights
The leading models and scores on this benchmark.
DocVQAtest Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for DocVQAtest
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is DocVQAtest?
What DocVQAtest measures and how its scores work.
DocVQAtest is a multimodal benchmark for answering questions about document images, including letters, memos, notes, and reports from the UCSF Industry Documents Library.
It measures performance in understanding document structure and content to answer questions, reported as Score in ratio units.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- DocVQAtest
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about DocVQAtest.
Which model scores highest on DocVQAtest?
Qwen3 VL 235B A22B Instruct is currently ranked first with 97.10%.
What are the top three models on DocVQAtest?
The current leaders are Qwen3 VL 235B A22B Instruct (97.10%), Qwen3 VL 32B Instruct (96.90%), and Qwen2-VL-72B-Instruct (96.50%).
Which DocVQAtest model has the lowest official input price?
No matched official input price is currently available.
Which models are fastest among DocVQAtest results?
The fastest matched records are Qwen3 VL 4B Thinking (8.77 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.