Benchmark Almanac

Encord Egocentric VQA Benchmark

GPT-6 Astra 76.5% leads since November 2026
Not in model cards

Models answer 1,500 multiple-choice questions about actions, objects, space and time in private first-person videos.

Encord’s November 11 leaderboard compares 18 model entries. GPT-6 Astra scored 76.51%.

Released
November 2026
Kind
General visual questions
Level
Everyday
Scored
Checked automatically
Answers
Multiple choice
Tasks
Kept private
Measures
Capability
Language
English
In and out
Video, Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
In views
Vision
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 21st, 2026
18
Entries on its own leaderboard, 11 models we track

GPT-6 Astra leads with 76.5% among models released by September 2026 GPT-6 Astra leads with 76.5% among models released by September 2026

Best score per entry from the benchmark's own leaderboard, read on November 21st, 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. This view shows only models we track, each at its best.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-6 Astra 76.5%
  2. 2 Gemini 3.7 Flash 73.3%
  3. 3 Gemini 3.8 Flash 73%
  4. 4 Qwen 3.8 27B 67.4%
  5. 5 Cosmos3-Super 66.8%
  6. 6 Gemini 3.1 Pro 66.2%
  7. 7 InternVL3-78B 58.2%
  8. 8 Qwen3-VL-4B 55.7%
  9. 9 Qwen3.6 27B 55.6%
  10. 10 InternVL3-38B 55.5%
  11. 11 LLaVA-OneVision2-8B 54.7%
  12. 12 PhysBrain (LIBERO) 53.5%
  13. 13 Molmo2-8B 52.8%
  14. 14 GPT-5.6 Luna 52.3%
  15. 15 InternVL3-8B 51.4%
  16. 16 Qwen3-VL-8B 50.1%
  17. 17 InternVideo3-8B 43.6%
  18. 18 Kimi-VL-A3B 23.1%

Source: Official Encord Egocentric VQA Benchmark leaderboard

#EntryScore ±SettingDateSource
1 GPT-6 Astra 76.5% ±2.1 Sep 2026 Official leaderboard
2 Gemini 3.7 Flash 73.3% ±2.2 Aug 2026 Official leaderboard
3 Gemini 3.8 Flash 73% ±2.3 Sep 2026 Official leaderboard
4 Qwen 3.8 27B 67.4% ±2.4 Aug 2026 Official leaderboard
5 Cosmos3-Super 66.8% ±2.4 Official leaderboard
6 Gemini 3.1 Pro 66.2% ±2.4 Feb 2026 Official leaderboard
7 InternVL3-78B 58.2% ±2.5 Apr 2025 Official leaderboard
8 Qwen3-VL-4B 55.7% ±2.5 Official leaderboard
9 Qwen3.6 27B 55.6% ±2.5 Apr 2026 Official leaderboard
10 InternVL3-38B 55.5% ±2.5 Apr 2025 Official leaderboard
11 LLaVA-OneVision2-8B 54.7% ±2.5 Official leaderboard
12 PhysBrain (LIBERO) 53.5% ±2.5 Official leaderboard
13 Molmo2-8B 52.8% ±2.5 Dec 2025 Official leaderboard
14 GPT-5.6 Luna 52.3% ±2.5 Jul 2026 Official leaderboard
15 InternVL3-8B 51.4% ±2.6 Apr 2025 Official leaderboard
16 Qwen3-VL-8B 50.1% ±2.5 Official leaderboard
17 InternVideo3-8B 43.6% ±2.5 Official leaderboard
18 Kimi-VL-A3B 23.1% ±2.1 Official leaderboard
#ModelScore ±SettingDateSource
1 GPT-6 Astra 76.5% ±2.1 Sep 2026 Official leaderboard
2 Gemini 3.7 Flash 73.3% ±2.2 Aug 2026 Official leaderboard
3 Gemini 3.8 Flash 73% ±2.3 Sep 2026 Official leaderboard
4 Qwen 3.8 27B 67.4% ±2.4 Aug 2026 Official leaderboard
5 Gemini 3.1 Pro 66.2% ±2.4 Feb 2026 Official leaderboard
6 InternVL3-78B 58.2% ±2.5 Apr 2025 Official leaderboard
7 Qwen3.6 27B 55.6% ±2.5 Apr 2026 Official leaderboard
8 InternVL3-38B 55.5% ±2.5 Apr 2025 Official leaderboard
9 Molmo2-8B 52.8% ±2.5 Dec 2025 Official leaderboard
10 GPT-5.6 Luna 52.3% ±2.5 Jul 2026 Official leaderboard
11 InternVL3-8B 51.4% ±2.6 Apr 2025 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI
Own leaderboard 1 leader
GPT-6 Astra 76.5%

Source: Official Encord Egocentric VQA Benchmark leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-6 Astra Gemini 3.1 Pro InternVL3-78B 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-6 Astra Gemini 3.1 Pro

Source: Official Encord Egocentric VQA Benchmark 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 Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (18 entries).
  • Pass Models scored 11 models we track, plus 7 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for GPT-6 Astra.
  • Note Frontier models Only 1 of the top 10 has a score. The strongest tested is GPT-6 Astra (#4).
  • Pass Headroom The best is 76.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on encord.com: 11 rows for models we track and 7 other entries at the last read.
  • Pass Last read November 21st, 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.

How we check