long context benchmark
EgoSchema Leaderboard
EgoSchema is a diagnostic benchmark for very long-form video language understanding with over 5000 human curated multiple choice questions based on 3-minute video clips from Ego4D.
Updated Sep 5, 2026
Models9
Model coverage9
MetricScore
EvidenceB
EgoSchema Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score77.90% | Percentile100.00% | Participants9 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score76.20% | Percentile87.50% | Participants9 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score72.20% | Percentile75.00% | Participants9 | EvidenceC | Evaluated |
| Rank04 | ModelAM Nova ProAmazon | Score72.10% | Percentile62.50% | Participants9 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score71.50% | Percentile50.00% | Participants9 | EvidenceC | Evaluated |
| Rank06 | ModelAMNova LiteAmazon | Score71.40% | Percentile37.50% | Participants9 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score68.60% | Percentile25.00% | Participants9 | EvidenceC | Evaluated |
| Rank08 | ModelGO | Score67.20% | Percentile12.50% | Participants9 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score55.70% | Percentile0.00% | Participants9 | EvidenceC | Evaluated |
EgoSchema Highlights
The leading models and scores on this benchmark.
EgoSchema Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for EgoSchema
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is EgoSchema?
What EgoSchema measures and how its scores work.
EgoSchema is a long_context benchmark covering natural human activities and behaviors in Ego4D video clips.
It measures very long-form video language understanding using multiple choice 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
- EgoSchema
- Modality
- video
- Primary category
- long context
- 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 EgoSchema.
Which model scores highest on EgoSchema?
Qwen2-VL-72B-Instruct is currently ranked first with 77.90%.
What are the top three models on EgoSchema?
The current leaders are Qwen2-VL-72B-Instruct (77.90%), Qwen2.5 VL 72B Instruct (76.20%), and GPT-4o (72.20%).
Which EgoSchema model has the lowest official input price?
Nova Lite has the lowest matched official input price at $0.06 input / $0.24 output per 1M tokens.
Which models are fastest among EgoSchema results?
The fastest matched records are Gemini 2.0 Flash (183.00 tok/s via Google), GPT-4o (132.00 tok/s via OpenAI), and Gemini 1.0 Pro (120.00 tok/s via Google).
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.