Benchmark Almanac

SkateBench v2

Gemini 3.1 Pro 96.9% leads since March 2026
Not in model cards

Models name skateboard tricks from technical trick definitions. The score is the share named correctly.

390 trick definitions, private after v1 saturated. The board lists 28 models but stopped near Gemini 3.1 Pro (97%), so current models are missing. Release month comes from secondary reports of the maker's post.

Released
March 2026
Kind
Trivia and open questions
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
Kept private
Strongest model tested
GPT-5.4 Pro #19 on the capability index
Last model added
March 2026
Scores refreshed
November 12th, 2026
21
Models scored on its own leaderboard

Gemini 3.1 Pro leads with 96.9% among models released by March 2026

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

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:

21 models. Scroll sideways for the rest.

  1. 96.9 Gemini 3.1 Pro
  2. 81.5 GPT-5.4
  3. 78.5 GPT-5.4 Pro
  4. 77.4 GPT-5
  5. 76.2 Gemini 3 Pro
  6. 76.2 Gemini 3 Flash
  7. 71.8 GPT-5.1
  8. 71.8 GPT-5.2 Pro
  9. 69.0 GPT-5.2
  10. 66.7 GLM-5
  11. 64.4 Claude Opus 4.6
  12. 61.3 Grok 4
  13. 57.2 Kimi K2.5
  14. 50.3 Claude Opus 4.5
  15. 46.7 DeepSeek-V3.2
  16. 43.9 Kimi K2 Thinking
  17. 38.0 GPT-5 mini
  18. 35.4 Grok 4.1 Fast
  19. 16.4 gpt-oss-120b
  20. 15.1 Claude Sonnet 4.6
  21. 13.8 MiniMax-M2.5
  1. 1 Gemini 3.1 Pro 96.9%
  2. 2 GPT-5.4 high 81.5%
  3. 3 GPT-5.4 Pro thinking 78.5%
  4. 4 GPT-5 high 77.4%
  5. 5 Gemini 3 Pro 76.2%
  6. 6 Gemini 3 Flash low 76.2%
  7. 7 GPT-5.1 71.8%
  8. 8 GPT-5.2 Pro 71.8%
  9. 9 GPT-5.2 xhigh 69%
  10. 10 GLM-5 66.7%
  11. 11 Claude Opus 4.6 high 64.4%
  12. 12 Grok 4 61.3%
  13. 13 Kimi K2.5 57.2%
  14. 14 Claude Opus 4.5 high 50.3%
  15. 15 DeepSeek-V3.2 high 46.7%
  16. 16 Kimi K2 Thinking 43.9%
  17. 17 GPT-5 mini 38%
  18. 18 Grok 4.1 Fast 35.4%
  19. 19 gpt-oss-120b high 16.4%
  20. 20 Claude Sonnet 4.6 15.1%
  21. 21 MiniMax-M2.5 13.8%

Source: Official SkateBench v2 leaderboard

#ModelScoreSettingDateSource
1 Gemini 3.1 Pro 96.9% Feb 2026 Official leaderboard
2 GPT-5.4 81.5% high Mar 2026 Official leaderboard
3 GPT-5.4 Pro 78.5% thinking Mar 2026 Official leaderboard
4 GPT-5 77.4% high Aug 2025 Official leaderboard
5 Gemini 3 Pro 76.2% Nov 2025 Official leaderboard
6 Gemini 3 Flash 76.2% low Dec 2025 Official leaderboard
7 GPT-5.1 71.8% Nov 2025 Official leaderboard
8 GPT-5.2 Pro 71.8% Dec 2025 Official leaderboard
9 GPT-5.2 69% xhigh Dec 2025 Official leaderboard
10 GLM-5 66.7% Feb 2026 Official leaderboard
11 Claude Opus 4.6 64.4% high Feb 2026 Official leaderboard
12 Grok 4 61.3% Jul 2025 Official leaderboard
13 Kimi K2.5 57.2% Jan 2026 Official leaderboard
14 Claude Opus 4.5 50.3% high Nov 2025 Official leaderboard
15 DeepSeek-V3.2 46.7% high Dec 2025 Official leaderboard
16 Kimi K2 Thinking 43.9% Nov 2025 Official leaderboard
17 GPT-5 mini 38% Aug 2025 Official leaderboard
18 Grok 4.1 Fast 35.4% Nov 2025 Official leaderboard
19 gpt-oss-120b 16.4% high Aug 2025 Official leaderboard
20 Claude Sonnet 4.6 15.1% Feb 2026 Official leaderboard
21 MiniMax-M2.5 13.8% Feb 2026 Official leaderboard

Value Matrix

Accuracy against average response time; the page toggles this view with cost, and top-left is best. The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.

0% 20% 40% 60% 80% 100% 110100 Average response time (s), log scale, lower is better Accuracy Grok 4: Accuracy 61.3%, average response time 60.9 s gpt-oss-120b (high): Accuracy 16.4%, average response time 32 s GPT-5 (high): Accuracy 77.4%, average response time 53.7 s GPT-5: Accuracy 75.4%, average response time 28.3 s GPT-5 mini: Accuracy 38%, average response time 15.5 s Kimi K2 Thinking: Accuracy 43.9%, average response time 25.3 s GPT-5.1: Accuracy 71.8%, average response time 25.6 s Gemini 3 Pro: Accuracy 76.2%, average response time 16.9 s Grok 4.1 Fast: Accuracy 35.4%, average response time 20.3 s Claude Opus 4.5 (high): Accuracy 50.3%, average response time 22.5 s DeepSeek-V3.2 (high): Accuracy 46.7%, average response time 51.3 s GPT-5.2 (xhigh): Accuracy 69%, average response time 36 s GPT-5.2 (high): Accuracy 44.6%, average response time 8.17 s GPT-5.2: Accuracy 41.5%, average response time 6.14 s GPT-5.2 Pro: Accuracy 71.8%, average response time 42.3 s Gemini 3 Flash (high): Accuracy 75.1%, average response time 11.1 s Kimi K2.5: Accuracy 57.2%, average response time 81.2 s Claude Opus 4.6 (high): Accuracy 64.4%, average response time 13.5 s GLM-5: Accuracy 66.7%, average response time 43.7 s MiniMax-M2.5: Accuracy 13.8%, average response time 11.6 s Claude Sonnet 4.6: Accuracy 15.1%, average response time 12.9 s GPT-5.4 (xhigh): Accuracy 81.3%, average response time 86.8 s GPT-5.4 Pro (thinking): Accuracy 78.5%, average response time 84.3 s GPT-5 (minimal): Accuracy 16.7%, average response time 1.17 s Gemini 3 Flash (low): Accuracy 76.2%, average response time 4.19 s Gemini 3.1 Pro: Accuracy 96.9%, average response time 22 s GPT-5.4 (high): Accuracy 81.5%, average response time 21.6 s GPT-5.4 (medium): Accuracy 78.5%, average response time 11.9 s Gemini 3.1 ProGPT-5.4 (high)GPT-5.4 (medium)Gemini 3 Flash (low)GPT-5 (minimal) 0% 20% 40% 60% 80% 100% 110100 Average response time (s), log scale, lower is better Accuracy Grok 4: Accuracy 61.3%, average response time 60.9 s gpt-oss-120b (high): Accuracy 16.4%, average response time 32 s GPT-5 (high): Accuracy 77.4%, average response time 53.7 s GPT-5: Accuracy 75.4%, average response time 28.3 s GPT-5 mini: Accuracy 38%, average response time 15.5 s Kimi K2 Thinking: Accuracy 43.9%, average response time 25.3 s GPT-5.1: Accuracy 71.8%, average response time 25.6 s Gemini 3 Pro: Accuracy 76.2%, average response time 16.9 s Grok 4.1 Fast: Accuracy 35.4%, average response time 20.3 s Claude Opus 4.5 (high): Accuracy 50.3%, average response time 22.5 s DeepSeek-V3.2 (high): Accuracy 46.7%, average response time 51.3 s GPT-5.2 (xhigh): Accuracy 69%, average response time 36 s GPT-5.2 (high): Accuracy 44.6%, average response time 8.17 s GPT-5.2: Accuracy 41.5%, average response time 6.14 s GPT-5.2 Pro: Accuracy 71.8%, average response time 42.3 s Gemini 3 Flash (high): Accuracy 75.1%, average response time 11.1 s Kimi K2.5: Accuracy 57.2%, average response time 81.2 s Claude Opus 4.6 (high): Accuracy 64.4%, average response time 13.5 s GLM-5: Accuracy 66.7%, average response time 43.7 s MiniMax-M2.5: Accuracy 13.8%, average response time 11.6 s Claude Sonnet 4.6: Accuracy 15.1%, average response time 12.9 s GPT-5.4 (xhigh): Accuracy 81.3%, average response time 86.8 s GPT-5.4 Pro (thinking): Accuracy 78.5%, average response time 84.3 s GPT-5 (minimal): Accuracy 16.7%, average response time 1.17 s Gemini 3 Flash (low): Accuracy 76.2%, average response time 4.19 s Gemini 3.1 Pro: Accuracy 96.9%, average response time 22 s GPT-5.4 (high): Accuracy 81.5%, average response time 21.6 s GPT-5.4 (medium): Accuracy 78.5%, average response time 11.9 s Gemini 3.1 ProGPT-5.4 (high)GPT-5.4 (medium)Gemini 3 Flash (low)GPT-5 (minimal)
ModelSettingAccuracyAverage response time
Gemini 3.1 Pro 96.9% 21.98 s
GPT-5.4 high 81.5% 21.62 s
GPT-5.4 xhigh 81.3% 86.81 s
GPT-5.4 medium 78.5% 11.94 s
GPT-5.4 Pro thinking 78.5% 84.29 s
GPT-5 high 77.4% 53.68 s
Gemini 3 Flash low 76.2% 4.19 s
Gemini 3 Pro 76.2% 16.94 s
GPT-5 75.4% 28.3 s
Gemini 3 Flash high 75.1% 11.11 s
GPT-5.1 71.8% 25.57 s
GPT-5.2 Pro 71.8% 42.27 s
GPT-5.2 xhigh 69% 36.04 s
GLM-5 66.7% 43.66 s
Claude Opus 4.6 high 64.4% 13.49 s
Grok 4 61.3% 60.92 s
Kimi K2.5 57.2% 81.24 s
Claude Opus 4.5 high 50.3% 22.51 s
DeepSeek-V3.2 high 46.7% 51.32 s
GPT-5.2 high 44.6% 8.17 s
Kimi K2 Thinking 43.9% 25.34 s
GPT-5.2 41.5% 6.14 s
GPT-5 mini 38% 15.52 s
Grok 4.1 Fast 35.4% 20.27 s
GPT-5 minimal 16.7% 1.17 s
gpt-oss-120b high 16.4% 32.02 s
Claude Sonnet 4.6 15.1% 12.9 s
MiniMax-M2.5 13.9% 11.59 s

The best score, 96.9%, costs $0.02 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI DeepSeek Google DeepMind Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.0001 $0.001 $0.01 $0.10 Cost per task, log scale Gemini 3.1 Pro GPT-5.4, high GPT-5.4, medium Gemini 3 Flash, low DeepSeek-V3.2, high GPT-5, minimal 0% 20% 40% 60% 80% 100% $0.0001 $0.001 $0.01 $0.10 Cost per task, log scale Gemini 3.1 Pro GPT-5.4, high

Source: Official SkateBench v2 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 3.1 Pro 96.9% $0.02 1,718
GPT-5.4 high 81.5% $0.02 1,180
GPT-5.4 xhigh 81.3% $0.08 5,336
GPT-5.4 medium 78.5% $0.009 574
GPT-5.4 Pro thinking 78.5% $0.08 5,134
GPT-5 high 77.4% $0.02 2,439
Gemini 3 Pro 76.2% $0.02 1,504
Gemini 3 Flash low 76.2% $0.001 437
GPT-5 75.4% $0.01 1,159
Gemini 3 Flash high 75.1% $0.005 1,603
GPT-5.1 71.8% $0.02 1,854
GPT-5.2 Pro 71.8% $0.07 386
GPT-5.2 xhigh 69% $0.03 2,113
GLM-5 66.7% $0.006 2,277
Claude Opus 4.6 high 64.4% $0.02 593
Grok 4 61.3% $0.03 2,260
Kimi K2.5 57.2% $0.006 2,064
Claude Opus 4.5 high 50.3% $0.03 1,219
DeepSeek-V3.2 high 46.7% $0.0008 1,398
GPT-5.2 high 44.6% $0.005 338
Kimi K2 Thinking 43.9% $0.003 1,260
GPT-5.2 41.5% $0.003 216
GPT-5 mini 38% $0.002 784
Grok 4.1 Fast 35.4% $0.001 1,939
GPT-5 minimal 16.7% $0.0004 23
gpt-oss-120b high 16.4% $0.001 2,175
Claude Sonnet 4.6 15.1% $0.01 655
MiniMax-M2.5 13.8% $0.0005 356

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 Passed 85% in Mar 2026, by Gemini 3.1 Pro
Gemini 3.1 Pro 96.9%

Source: Official SkateBench v2 leaderboard

Gemini 3.1 Pro on the official leaderboard passed 85% in February 2026

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

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro GPT-5 Grok 4 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro GPT-5

Source: Official SkateBench v2 leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Knowledge

All 121 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • KMMLU-Pro In use Models answer multiple-choice questions taken from Korean National Professional Licensure exams. It tests professional knowledge used in Korean industry. Solar Pro 4 79.2%
  • HLE-Verified In use Verified and revised version of Humanity's Last Exam (1,811 items) with uncertain items removed. Seed2.0 Pro 73.6%