Benchmark Almanac

IntPhys2-Timing

Gemini 3.8 Flash 77.4% leads since October 2026
Not in model cards

Models identify when a physical rule breaks in a video. The authors add violation intervals to IntPhys2 and score temporal localization.

Made by
Taekoan Yoo and colleagues
Released
October 2026
Kind
Physical prediction
Level
Everyday
Scored
Checked automatically
Answers
Prediction
Tasks
All public
Measures
Capability
In and out
Video → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Based on
IntPhys 2
In views
Vision
Strongest model tested
Gemini 3.8 Flash #24 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
5
Entries in its paper, 3 models we track

Gemini 3.8 Flash leads with 77.4% among models released by September 2026 Gemini 3.8 Flash leads with 77.4% among models released by September 2026

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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 Gemini 3.8 Flash 77.4%
  2. 2 Qwen 3.8 27B 73.1%
  3. 3 GPT-5.4 70.3%
  4. 4 V-JEPA2-g 61.1%
  5. 5 V-JEPA2-h 44.9%

Source: The IntPhys2-Timing paper

Source: The IntPhys2-Timing paper

#EntryScoreSettingDateSource
1 Gemini 3.8 Flash 77.4% PhysTrace Sep 2026 Official leaderboard
2 Qwen 3.8 27B 73.1% PhysTrace Aug 2026 Official leaderboard
3 GPT-5.4 70.3% PhysTrace Mar 2026 Official leaderboard
4 V-JEPA2-g 61.1% VoE Official leaderboard
5 V-JEPA2-h 44.9% VoE Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3.8 Flash 77.4% PhysTrace Sep 2026 Official leaderboard
2 Qwen 3.8 27B 73.1% PhysTrace Aug 2026 Official leaderboard
3 GPT-5.4 70.3% PhysTrace Mar 2026 Official leaderboard

Rankings by part

We read 20 parts from the benchmark’s paper, each ranking models on one part of the test, such as Continuity: Easy, Continuity: Medium and Continuity: Hard. Gemini 3.8 Flash leads 9 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.

Continuity: Easy

3 models we track have a score here.

  1. 1 GPT-5.4 ↑2 87.5%
  2. 2 Gemini 3.8 Flash ↓1 87.5%
  3. 3 Qwen 3.8 27B ↓1 75%

Number one over time

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

Google DeepMind
Benchmark paper Not tested since Oct 2026

Source: The IntPhys2-Timing paper

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
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.8 Flash Qwen 3.8 27B GPT-5.4 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.8 Flash GPT-5.4

Source: Official IntPhys2-Timing 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 World models

All 53 in World models
  • AgentWorldBench In use Language world models predict the next environment observation after an agent action, in 7 domains from MCP and search to Android and desktop OS. A judge model scores each prediction against the real observation on five dimensions. GPT-5.4 58.3
  • IntPhys 2 In use Models watch simulated videos and judge whether each event is physically possible, testing object permanence, immutability, continuity and solidity. Scored by accuracy. Muse Spark Open 68.3%
  • QwenWorldBench Fading Models act as world models that simulate agent environments in 7 domains, such as terminal, web, and Android. A rubric judge scores five dimensions. Qwen3.7-Plus 62.1%
  • VPCT Fading Models view an image of a ball, ramps, and three buckets and predict which bucket the ball falls into. Scored by accuracy on 100 problems. Gemini 3 Pro 91%
  • ContPhy Fading Models infer physical properties such as mass and density of fluids and soft bodies from videos and predict their dynamics. Seed2.0 Pro 67.4%
  • RBench Only in system cards 650 image+text prompts for robot video generation (tasks and embodiments). MLLM judge scores task completion, physical plausibility, stability. Wan 2.6 60.7%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (5 entries).
  • Pass Models scored 3 models we track, plus 2 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for Gemini 3.8 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.8 Flash (#24).
  • Pass Headroom The best is 77.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 3 rows for models we track and 2 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