Benchmark Almanac

ReactionBench

GPT-5 0.51 leads since October 2026
Not in model cards

Models rank a viewer’s reactions, match a reaction to one of four video clips, and explain which cue prompted it.

The authors report seven models. They score intensity by rank correlation, matching by accuracy, and explanations with a judge. Researchers need approval for the videos.

Released
October 2026
Kind
Video understanding
Level
Everyday
Scored
Checked automatically
Answers
Prediction
Tasks
All public
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
Anyone
In views
Vision
Strongest model tested
Gemini 3.7 Flash #26 on the capability index
Last model added
August 2026
Scores refreshed
November 20th, 2026
7
Models scored on its own leaderboard

GPT-5 leads with 0.51 among models released by August 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. The board reports no overall score and ranks by stimulus-reaction matching accuracy, measured from 0 to 1. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in August 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-5 Zero-shot 0.51
  2. 2 Gemini 3.7 Flash Zero-shot 0.50
  3. 3 Qwen 3.6 35B-A3B Zero-shot 0.48
  4. 4 Qwen3.6 27B Zero-shot 0.45
  5. 5 Qwen3.5 397B-A17B Zero-shot 0.41
  6. 6 Kimi K2.6 Zero-shot 0.38
  7. 7 Qwen3.5-9B Zero-shot 0.26

Source: Official ReactionBench leaderboard

#ModelScoreSettingDateSource
1 GPT-5 0.51 Zero-shot Aug 2025 Official leaderboard
2 Gemini 3.7 Flash 0.50 Zero-shot Aug 2026 Official leaderboard
3 Qwen 3.6 35B-A3B 0.48 Zero-shot Apr 2026 Official leaderboard
4 Qwen3.6 27B 0.45 Zero-shot Apr 2026 Official leaderboard
5 Qwen3.5 397B-A17B 0.41 Zero-shot Feb 2026 Official leaderboard
6 Kimi K2.6 0.38 Zero-shot Apr 2026 Official leaderboard
7 Qwen3.5-9B 0.26 Zero-shot Feb 2026 Official leaderboard

Rankings by part

We read 8 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as IP joint: ρ₀, IP joint: r and IP indep.: ρ₀. Gemini 3.7 Flash leads 4 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

IP joint: ρ₀

7 models we track have a score here.

  1. 1 Gemini 3.7 Flash ↑1 0.18
  2. 2 Qwen3.6 27B ↑2 0.16
  3. 3 Kimi K2.6 ↑3 0.09
  4. 4 Qwen3.5 397B-A17B ↑1 0.07
  5. 5 Qwen 3.6 35B-A3B ↓2 0.04
  6. 6 Qwen3.5-9B ↑1 0.03
  7. 7 GPT-5 ↓6 0.03

Number one over time

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

OpenAI
Own leaderboard Not tested since Oct 2026
GPT-5 0.51

Source: Official ReactionBench 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
0.30 0.40 0.50 2026 0.30 0.40 0.50 2026

Source: Official ReactionBench 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

All 695 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 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (7 entries).
  • Pass Models scored 7 models we track.
  • Pass Newest score The newest is from August 2026, for Gemini 3.7 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.7 Flash (#26).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on yibozh.github.io: 7 rows for models we track 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.

How we check