Benchmark Almanac

VideoScore-Bench

VideoScore (reg) 69.6 tops its leaderboard Gemini 1.5 Pro May 2024 39.7 leads the models we track
Not in model cards
Version 1

Models rate generated videos or choose which video better matches a human preference. The four test sets compare model judgments with human judgments.

The EVA-RSI team compared 21 evaluators on dynamic degree using 80 videos from the original test.

Made by
TIGER Lab and collaborators
Released
June 2024
Kind
Video understanding
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Video, Text → Text
Interaction
Single turn
Items from
Generated
Run by
Anyone
In views
Vision
Strongest model tested
GPT-4o May 2024 #162 on the capability index
Last model added
May 2024
Scores refreshed
November 20th, 2026
17
Entries on its own leaderboard, 6 models we track

VideoScore (reg) tops the leaderboard with 69.6 Gemini 1.5 Pro leads with 39.7 among models released by May 2024

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. The best model we track is Gemini 1.5 Pro, at 39.7. This view shows only models we track, each at its best.

The newest model with a score here came out in May 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 VideoScore (reg) 69.6
  2. 2 VideoScore-(gen) 55.6
  3. 3 Gemini 1.5 Pro May 2024 39.7
  4. 4 Gemini 1.5 Flash May 2024 39.4
  5. 5 GPT-4o May 2024 38.9
  6. 6 CLIP-sim 31.7
  7. 7 DINO-sim 30.3
  8. 8 SSIM-sim 29.5
  9. 9 CLIP-Score 28.6
  10. 10 llava-v1.5-7b 27.1
  11. 11 llava-v1.6-vicuna-7b 23.3
  12. 12 X-CLIP-Score 23.2
  13. 13 PIQE 19.6
  14. 14 BRISQUE 19.0
  15. 15 IDEFICS-9B-Instruct 18.3
  16. 16 MSE-dyn 10.6
  17. 17 SSIM-dyn 9.20

Source: Official VideoScore-Bench leaderboard

#EntryScoreSettingDateSource
1 VideoScore (reg) 69.6 Official leaderboard
2 VideoScore-(gen) 55.6 Official leaderboard
3 Gemini 1.5 Pro May 2024 39.7 May 2024 Official leaderboard
4 Gemini 1.5 Flash May 2024 39.4 May 2024 Official leaderboard
5 GPT-4o May 2024 38.9 May 2024 Official leaderboard
6 CLIP-sim 31.7 Official leaderboard
7 DINO-sim 30.3 Official leaderboard
8 SSIM-sim 29.5 Official leaderboard
9 CLIP-Score 28.6 Official leaderboard
10 llava-v1.5-7b 27.1 Oct 2023 Official leaderboard
11 llava-v1.6-vicuna-7b 23.3 Jan 2024 Official leaderboard
12 X-CLIP-Score 23.2 Official leaderboard
13 PIQE 19.6 Official leaderboard
14 BRISQUE 19.0 Official leaderboard
15 IDEFICS-9B-Instruct 18.3 Jul 2023 Official leaderboard
16 MSE-dyn 10.6 Official leaderboard
17 SSIM-dyn 9.20 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 1.5 Pro May 2024 39.7 May 2024 Official leaderboard
2 Gemini 1.5 Flash May 2024 39.4 May 2024 Official leaderboard
3 GPT-4o May 2024 38.9 May 2024 Official leaderboard
4 llava-v1.5-7b 27.1 Oct 2023 Official leaderboard
5 llava-v1.6-vicuna-7b 23.3 Jan 2024 Official leaderboard
6 IDEFICS-9B-Instruct 18.3 Jul 2023 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as VideoFeedback-test, EvalCrafter and GenAI-Bench. GPT-4o leads 2 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

VideoFeedback-test

6 models we track have a score here.

  1. 1 GPT-4o May 2024 ↑2 23.1
  2. 2 Gemini 1.5 Pro May 2024 ↓1 22.1
  3. 3 Gemini 1.5 Flash May 2024 ↓1 20.8
  4. 4 llava-v1.5-7b 8.50
  5. 5 IDEFICS-9B-Instruct ↑1 6.50
  6. 6 llava-v1.6-vicuna-7b ↓1 -3.10

Number one over time

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

Google DeepMind
Own leaderboard Not tested since Jun 2024

Source: Official VideoScore-Bench 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
15.0 20.0 25.0 30.0 35.0 40.0 2024 15.0 20.0 25.0 30.0 35.0 40.0 2024

Source: Official VideoScore-Bench 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.

Versions

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 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 official leaderboard (17 entries).
  • Pass Models scored 6 models we track, plus 11 other entries on its leaderboard.
  • Note Newest score The newest came 2 years ago, in May 2024, for Gemini 1.5 Flash. That’s expected for a test replaced by VideoScore-Bench v2.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-4o (#162).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on tiger-ai-lab.github.io: 6 rows for models we track and 11 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.

How we check