Benchmark
DocVQAtest Benchmark
What is DocVQAtest?
DocVQA is a Visual Question Answering benchmark on document images containing 50,000 questions defined on 12,000+ document images. The benchmark focuses on…
- Modality
- multimodal
- Evaluates
- multimodal, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 11
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/docvqatest?top_n=500
DocVQAtest results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Qwen2-VL-72B-Instruct | Qwen | 0.965 | 2024-08-29 | evidence |
| Qwen3 VL 235B A22B Instruct | Qwen | 0.971 | 2025-09-22 | evidence |
| Qwen3 VL 235B A22B Thinking | Qwen | 0.965 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Instruct | Qwen | 0.95 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Thinking | Qwen | 0.95 | 2025-09-22 | evidence |
| Qwen3 VL 32B Instruct | Qwen | 0.969 | 2025-09-22 | evidence |
| Qwen3 VL 32B Thinking | Qwen | 0.961 | 2025-09-22 | evidence |
| Qwen3 VL 4B Instruct | Qwen | 0.953 | 2025-09-22 | evidence |
| Qwen3 VL 4B Thinking | Qwen | 0.942 | 2025-09-22 | evidence |
| Qwen3 VL 8B Instruct | Qwen | 0.961 | 2025-09-22 | evidence |
| Qwen3 VL 8B Thinking | Qwen | 0.953 | 2025-09-22 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.