NEST
Models interpret children's gaze, gestures, and physical contact in social videos. The benchmark checks classification and choice questions and separately grades written descriptions.
- Made by
- Stanford University and collaborators
- Released
- November 2026
- 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
- Strongest model tested
- GPT-5 mini #92 on the capability index
- Last model added
- March 2026
- Scores refreshed
- November 20th, 2026
Qwen3.5-27B leads with 76% among models released by March 2026 Qwen3.5-27B leads with 76% among models released by March 2026
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. This view shows only models we track, each at its best.
The newest model with a score here came out in March 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 76.0
Qwen3.5-27B
- 73.9
Qwen3.5-9B
- 73.4 GPT-5 mini
- 69.7
Qwen3.5-4B
- 69.7
Gemini 3.1 Flash-Lite
- 69.6
Molmo2-8B
- 67.6
MiniCPM-V 4.5
- 66.4 Qwen3-VL 32B
- 63.4 openPangu-VL 7B
- 59.6
LLaVA-OneVision 7B
- 59.1 Qwen3-VL 8B
- 57.4
InternVL3.5-38B
- 55.5 Qwen3-VL 4B
- 54.8
Keye-VL-1.5-8B
- 46.1
InternVL3.5-8B
- 44.0 LLaVA-NeXT-Video 7B
- 41.9
InternVL3.5-4B
- 1
Qwen3.5-27B 76%
- 2
Qwen3.5-9B 73.9%
- 3 GPT-5 mini 73.4%
- 4
Qwen3.5-4B 69.7%
- 5
Gemini 3.1 Flash-Lite 69.7%
- 6
Molmo2-8B 69.6%
- 7
MiniCPM-V 4.5 67.6%
- 8 Qwen3-VL 32B 66.4%
- 9 openPangu-VL 7B 63.4%
- 10
LLaVA-OneVision 7B 59.6%
- 11 Qwen3-VL 8B 59.1%
- 12
InternVL3.5-38B 57.4%
- 13 Qwen3-VL 4B 55.5%
- 14
Keye-VL-1.5-8B 54.8%
- 15
InternVL3.5-8B 46.1%
- 16 LLaVA-NeXT-Video 7B 44%
- 17
InternVL3.5-4B 41.9%
Source: Official NEST leaderboard
- 1
Qwen3.5-27B 76%
- 2
Qwen3.5-9B 73.9%
- 3 GPT-5 mini 73.4%
- 4
Qwen3.5-4B 69.7%
- 5
Gemini 3.1 Flash-Lite 69.7%
- 6
Molmo2-8B 69.6%
- 7
MiniCPM-V 4.5 67.6%
- 8
LLaVA-OneVision 7B 59.6%
- 9
InternVL3.5-38B 57.4%
- 10
Keye-VL-1.5-8B 54.8%
- 11
InternVL3.5-8B 46.1%
- 12
InternVL3.5-4B 41.9%
Source: Official NEST leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 76% | Feb 2026 | Official leaderboard | |
| 2 | | 73.9% | Feb 2026 | Official leaderboard | |
| 3 | GPT-5 mini | 73.4% | Aug 2025 | Official leaderboard | |
| 4 | | 69.7% | Feb 2026 | Official leaderboard | |
| 5 | | 69.7% | Mar 2026 | Official leaderboard | |
| 6 | | 69.6% | Dec 2025 | Official leaderboard | |
| 7 | | 67.6% | Aug 2025 | Official leaderboard | |
| 8 | Qwen3-VL 32B 32B | 66.4% | Official leaderboard | ||
| 9 | openPangu-VL 7B 7B | 63.4% | Official leaderboard | ||
| 10 | | 59.6% | Aug 2024 | Official leaderboard | |
| 11 | Qwen3-VL 8B 8B | 59.1% | Official leaderboard | ||
| 12 | | 57.4% | Aug 2025 | Official leaderboard | |
| 13 | Qwen3-VL 4B 4B | 55.5% | Official leaderboard | ||
| 14 | | 54.8% | Aug 2025 | Official leaderboard | |
| 15 | | 46.1% | Aug 2025 | Official leaderboard | |
| 16 | LLaVA-NeXT-Video 7B 7B | 44% | Official leaderboard | ||
| 17 | | 41.9% | Aug 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 76% | Feb 2026 | Official leaderboard | |
| 2 | | 73.9% | Feb 2026 | Official leaderboard | |
| 3 | GPT-5 mini | 73.4% | Aug 2025 | Official leaderboard | |
| 4 | | 69.7% | Feb 2026 | Official leaderboard | |
| 5 | | 69.7% | Mar 2026 | Official leaderboard | |
| 6 | | 69.6% | Dec 2025 | Official leaderboard | |
| 7 | | 67.6% | Aug 2025 | Official leaderboard | |
| 8 | | 59.6% | Aug 2024 | Official leaderboard | |
| 9 | | 57.4% | Aug 2025 | Official leaderboard | |
| 10 | | 54.8% | Aug 2025 | Official leaderboard | |
| 11 | | 46.1% | Aug 2025 | Official leaderboard | |
| 12 | | 41.9% | Aug 2025 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Contextual, Behavioral and Interpersonal. Each ranks models on one part of the test. Qwen3.5-27B leads 2 of them, and GPT-5 mini leads the other one. Arrows show how far a model moved from its overall rank among the models we track.
Contextual
12 models we track have a score here.
- 1
Qwen3.5-27B 88.5%
- 2
Qwen3.5-4B ↑2 85.5%
- 3
Qwen3.5-9B ↓1 85.2%
- 4
Gemini 3.1 Flash-Lite ↑1 83.2%
- 5 GPT-5 mini ↓2 82.6%
- 6
MiniCPM-V 4.5 ↑1 82.6%
- 7
Molmo2-8B ↓1 81.2%
- 8
LLaVA-OneVision 7B 72.7%
- 9
InternVL3.5-38B 65.4%
- 10
Keye-VL-1.5-8B 60.8%
Behavioral
12 models we track have a score here.
- 1
Qwen3.5-27B 73.7%
- 2
Qwen3.5-9B 69.4%
- 3
Gemini 3.1 Flash-Lite ↑2 69.4%
- 4 GPT-5 mini ↓1 69.3%
- 5
Qwen3.5-4B ↓1 67.6%
- 6
Molmo2-8B 66.9%
- 7
MiniCPM-V 4.5 62.6%
- 8
InternVL3.5-38B ↑1 56.1%
- 9
LLaVA-OneVision 7B ↓1 52.3%
- 10
Keye-VL-1.5-8B 48.7%
Interpersonal
12 models we track have a score here.
- 1 GPT-5 mini ↑2 68.4%
- 2
Qwen3.5-9B 67%
- 3
Qwen3.5-27B ↓2 65.8%
- 4
Molmo2-8B ↑2 60.6%
- 5
MiniCPM-V 4.5 ↑2 57.7%
- 6
Gemini 3.1 Flash-Lite ↓1 56.5%
- 7
Qwen3.5-4B ↓3 56.1%
- 8
Keye-VL-1.5-8B ↑2 54.9%
- 9
LLaVA-OneVision 7B ↓1 53.6%
- 10
InternVL3.5-38B ↓1 50.9%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official NEST 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 NEST 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 8 months ago, in March 2026, for Gemini 3.1 Flash-Lite. 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 (17 entries).
- Pass Models scored 12 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 8 months ago, in March 2026, for Gemini 3.1 Flash-Lite.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 mini (#92).
- Pass Headroom The best is 76%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on laubravo.github.io: 12 rows 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.