HD-EPIC VQA
Models watch first-person kitchen videos and choose among five answers. Questions ask what people cooked, where they moved objects, and where they looked.
The organizers published 26,650 questions. Huawei’s team reported 41.6% accuracy with a system based on Qwen2.5-VL.
- Made by
- University of Bristol and the HD-EPIC team
- Released
- February 2025
- Kind
- Video understanding
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Another benchmark, Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Last model added
- January 2025
- Scores refreshed
- November 20th, 2026
Politecnico di Milano (dav) tops the leaderboard with 44.2% Qwen2.5-VL-7B leads with 28.7%
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen2.5-VL-7B, at 28.7%. This view shows only models we track, each at its best.
- 1 Politecnico di Milano (dav) 44.2%
- 2 Huawei Noah’s Ark Lab (ShaunYoung) 41.6%
- 3 LLavaVideo-72B Zero-shot 32.3%
- 4 Qwen2.5VL-32B-72B Zero-shot 32.1%
- 5 LLavaVideo-7B Zero-shot 30.2%
- 6
Qwen2.5-VL-7B Zero-shot 28.7%
Source: Official HD-EPIC VQA leaderboard
Source: Official HD-EPIC VQA leaderboard
| # | Entry | Score | Setting | Submitted by | Date | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | Politecnico di Milano (dav) | 44.2% | Politecnico di Milano (dav) | hd-epic.github.io | Official leaderboard | ||
| 2 | Huawei Noah’s Ark Lab (ShaunYoung) | 41.6% | Huawei Noah’s Ark Lab (ShaunYoung) | hd-epic.github.io | Official leaderboard | ||
| 3 | LLavaVideo-72B | 32.3% | Zero-shot | Ricoh Software Research Center | hd-epic.github.io | Official leaderboard | |
| 4 | Qwen2.5VL-32B-72B | 32.1% | Zero-shot | Ricoh Software Research Center | hd-epic.github.io | Official leaderboard | |
| 5 | LLavaVideo-7B | 30.2% | Zero-shot | Ricoh Software Research Center | hd-epic.github.io | Official leaderboard | |
| 6 | | 28.7% | Zero-shot | Ricoh Software Research Center | Jan 2025 | hd-epic.github.io | Official leaderboard |
| # | Model | Score | Setting | Submitted by | Date | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | | 28.7% | Zero-shot | Ricoh Software Research Center | Jan 2025 | hd-epic.github.io | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official HD-EPIC VQA leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 22 months ago, in January 2025, for Qwen2.5-VL-7B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (6 entries).
- Pass Models scored 1 model we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 22 months ago, in January 2025, for Qwen2.5-VL-7B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 28.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on codabench.org: 1 row for models we track and 5 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.