EgoCross
Models answer questions about events in first-person videos. The challenge scores accuracy on 957 four-choice questions across four domains.
The challenge organizers reported 19 open-source teams and 38 source-limited teams. OmniEgo used Qwen3-VL-4B and placed second in both tracks. Downloading the answer file requires a Hugging Face account.
- Made by
- Yanjun Li and colleagues
- Released
- August 2025
- Kind
- Video understanding
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
DomainWiseInfer tops the leaderboard with 67%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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.
- 1 DomainWiseInfer Source-Limited Track 67%
- 2 OmniEgo-R2 Open-Source Track 66.8%
- 3 Reflective Dialogue Open-Source Track 65.9%
- 4 anyone Open-Source Track 65.7%
- 5 cola_lover Open-Source Track 62.4%
- 6 zhengsr Open-Source Track 61.5%
- 7 handsomeboy_hun Open-Source Track 60.7%
- 8 TokenInj-RAA Source-Limited Track 60.6%
- 9 dghf2022 Source-Limited Track 58.9%
- 10 zr-zhou Source-Limited Track 58.1%
- 11 Ego101 Source-Limited Track 58%
- 12 xiang_li Source-Limited Track 58%
- 13 lalala111 Source-Limited Track 57.4%
- 14 odsigjoasd Source-Limited Track 56.1%
Source: The EgoCross paper
| # | Entry | Score | Setting | Submitted by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | DomainWiseInfer | 67% | Source-Limited Track | DomainWiseInfer | Official leaderboard | |
| 2 | OmniEgo-R2 | 66.8% | Open-Source Track | OmniEgo-R2 | Official leaderboard | |
| 3 | Reflective Dialogue | 65.9% | Open-Source Track | Reflective Dialogue | Official leaderboard | |
| 4 | anyone | 65.7% | Open-Source Track | anyone | Official leaderboard | |
| 5 | cola_lover | 62.4% | Open-Source Track | cola_lover | Official leaderboard | |
| 6 | zhengsr | 61.5% | Open-Source Track | zhengsr | Official leaderboard | |
| 7 | handsomeboy_hun | 60.7% | Open-Source Track | handsomeboy_hun | Official leaderboard | |
| 8 | TokenInj-RAA | 60.6% | Source-Limited Track | TokenInj-RAA | Official leaderboard | |
| 9 | dghf2022 | 58.9% | Source-Limited Track | dghf2022 | Official leaderboard | |
| 10 | zr-zhou | 58.1% | Source-Limited Track | zr-zhou | Official leaderboard | |
| 11 | Ego101 | 58% | Source-Limited Track | Ego101 | Official leaderboard | |
| 12 | xiang_li | 58% | Source-Limited Track | xiang_li | Official leaderboard | |
| 13 | lalala111 | 57.4% | Source-Limited Track | lalala111 | Official leaderboard | |
| 14 | odsigjoasd | 56.1% | Source-Limited Track | odsigjoasd | Official 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
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (14 entries).
- Pass Models scored 14 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom No scores yet.
Leaderboard import 3 pass
- Pass Importer We read the board on egocross-benchmark.github.io: 14 entries at the last read, none a model we track.
- 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.