Visual Reasoning Benchmark (Fab AI)
Models choose which picture completes a pattern, matches another shape, or shows a folded figure. Questions come from school exams in Zambia and India.
Fab AI’s leaderboard scores 67 model configurations. The authors describe 701 questions and publish example puzzles and scoring code. They keep the full exam images and answer keys private.
- Made by
- Fab AI
- Released
- September 2025
- Kind
- Spatial reasoning and counting
- Level
- School
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Exams and contests
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision
- Strongest model tested
- GPT-5.6 Sol #14 on the capability index
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
Gemini 3.7 Flash leads with 89.9% among models released by August 2026 Gemini 3.7 Flash leads with 89.9% among models released by August 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. Every run is in the next chart. 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 August 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 15 more
- 89.9
Gemini 3.7 Flash
- 86.0
Gemini 3.5 Flash
- 85.7
Gemini 3.6 Flash
- 84.9 GPT-5.6 Sol
- 81.3
Gemini 3.1 Pro
- 78.0
Gemini 3 Flash
- 77.7 GPT-5.5
- 76.3 GPT-5.4
- 76.0
Qwen3.5 397B-A17B
- 76.0 Kimi K3
- 75.9
Gemini 3 Pro
- 73.3
Qwen 3.6 Plus
- 72.6
Qwen3.5-Plus
- 67.3 GPT-5.2
- 66.2
Qwen3.6 27B
- 64.2 GPT-5.4 Mini
- 64.2
Qwen 3.6 35B-A3B
- 63.5
Qwen 3.6 Flash
- 61.6
Qwen3.5-Flash
- 60.5 Kimi K2.5
- 58.1
Gemini 2.5 Pro
- 56.9
Gemini 3.1 Flash-Lite
- 51.5 GPT-5
- 50.6
Gemini 2.5 Flash
- 49.1 GPT-5 mini
- 47.6 GPT-5.1
- 46.6 Grok 4.20
- 46.4 o3
- 46.2 Grok 4.3
- 45.4
Qwen3-VL-30B-A3B-Thinking
- 44.5 GLM-4.6V
- 43.2 o4-mini
- 41.7 GPT-4.1 mini
- 41.2 Claude Opus 4
- 39.7 Doubao-1.5 Vision Pro
- 39.1 GPT-4.1
- 38.7 Claude 3.7 Sonnet
- 38.7 Grok 4
- 38.2
Gemini 2.5 Flash-Lite
- 38.1
Gemini 2.0 Flash
- 37.7 GLM-4.1V-9B-Thinking
- 35.2
Llama 4 Maverick
- 35.0 GPT-5.4 Nano
- 34.7 Claude Sonnet 4
- 33.2 GPT-4o
- 32.1
Pixtral Large
- 31.2
Llama 4 Scout
- 31.1 GPT-5 nano
- 30.0
Qwen2.5-VL-72B
- 29.2
Gemma 3 12B
- 28.8 GPT-4o mini
- 28.8
Mistral Medium 3
- 28.7
Gemma 3 27B
- 28.4 Grok 2 Vision (1212)
- 27.4 GPT-4.1 nano
- 27.2 Claude 3.5 Haiku
- 26.8
Mistral Small 3.1
- 26.0 LFM2.5 VL 1.6B
- 23.0
Gemma 3 4B
- 1
Gemini 3.7 Flash 89.9%
- 2
Gemini 3.5 Flash 86%
- 3
Gemini 3.6 Flash 85.7%
- 4 GPT-5.6 Sol 84.9%
- 5
Gemini 3.1 Pro 81.3%
- 6
Gemini 3 Flash 78%
- 7 GPT-5.5 77.7%
- 8 GPT-5.4 medium 76.3%
- 9
Qwen3.5 397B-A17B 76%
- 10 Kimi K3 76%
- 11
Gemini 3 Pro 75.9%
- 12
Qwen 3.6 Plus 73.3%
- 13
Qwen3.5-Plus 72.6%
- 14 GPT-5.2 medium 67.3%
- 15
Qwen3.6 27B 66.2%
- 16 GPT-5.4 Mini medium 64.2%
- 17
Qwen 3.6 35B-A3B 64.2%
- 18
Qwen 3.6 Flash 63.5%
- 19
Qwen3.5-Flash 61.6%
- 20 Kimi K2.5 60.5%
- 21
Gemini 2.5 Pro Jun 2025 58.1%
- 22
Gemini 3.1 Flash-Lite 56.9%
- 23 GPT-5 medium 51.5%
- 24
Gemini 2.5 Flash Sep 2025 50.6%
- 25 GPT-5 mini medium 49.1%
- 26 GPT-5.1 medium 47.6%
- 27 Grok 4.20 reasoning 46.6%
- 28 o3 46.4%
- 29 Grok 4.3 46.2%
- 30
Qwen3-VL-30B-A3B-Thinking 45.4%
- 31 GLM-4.6V 44.5%
- 32 o4-mini 43.2%
- 33 GPT-4.1 mini 41.7%
- 34 Claude Opus 4 Thinking (low) 41.2%
- 35 Doubao-1.5 Vision Pro 39.7%
- 36 GPT-4.1 39.1%
- 37 Claude 3.7 Sonnet 38.7%
- 38 Grok 4 38.7%
- 39
Gemini 2.5 Flash-Lite Sep 2025 38.2%
- 40
Gemini 2.0 Flash Feb 2025 38.1%
- 41 GLM-4.1V-9B-Thinking 37.7%
- 42
Llama 4 Maverick 35.2%
- 43 GPT-5.4 Nano medium 35%
- 44 Claude Sonnet 4 34.7%
- 45 GPT-4o Mar 2025 33.2%
- 46
Pixtral Large 32.1%
- 47
Llama 4 Scout 31.2%
- 48 GPT-5 nano medium 31.1%
- 49
Qwen2.5-VL-72B 30%
- 50
Gemma 3 12B 29.2%
- 51 GPT-4o mini 28.8%
- 52
Mistral Medium 3 28.8%
- 53
Gemma 3 27B 28.7%
- 54 Grok 2 Vision (1212) 28.4%
- 55 GPT-4.1 nano 27.4%
- 56 Claude 3.5 Haiku 27.2%
- 57
Mistral Small 3.1 26.8%
- 58 LFM2.5 VL 1.6B 26%
- 59
Gemma 3 4B 23%
Source: Official Visual Reasoning Benchmark (Fab AI) leaderboard
- 89.9
Gemini 3.7 Flash
- 86.0
Gemini 3.5 Flash
- 85.7
Gemini 3.6 Flash
- 84.9 GPT-5.6 Sol
- 81.3
Gemini 3.1 Pro
- 78.0
Gemini 3 Flash
- 77.7 GPT-5.5
- 76.3 GPT-5.4
- 76.0
Qwen3.5 397B-A17B
- 76.0 Kimi K3
- 75.9
Gemini 3 Pro
- 73.3
Qwen 3.6 Plus
- 72.6
Qwen3.5-Plus
- 67.3 GPT-5.2
- 66.2
Qwen3.6 27B
- 64.2 GPT-5.4 Mini
- 64.2
Qwen 3.6 35B-A3B
- 63.5
Qwen 3.6 Flash
- 61.6
Qwen3.5-Flash
- 60.5 Kimi K2.5
- 58.1
Gemini 2.5 Pro
- 56.9
Gemini 3.1 Flash-Lite
- 51.5 GPT-5
- 50.6
Gemini 2.5 Flash
- 49.1 GPT-5 mini
- 47.6 GPT-5.1
- 46.6 Grok 4.20
- 46.4 o3
- 46.2 Grok 4.3
- 45.4
Qwen3-VL-30B-A3B-Thinking
- 44.5 GLM-4.6V
- 43.2 o4-mini
- 41.7 GPT-4.1 mini
- 41.2 Claude Opus 4
- 39.1 GPT-4.1
- 38.7 Claude 3.7 Sonnet
- 38.7 Grok 4
- 38.2
Gemini 2.5 Flash-Lite
- 38.1
Gemini 2.0 Flash
- 37.7 GLM-4.1V-9B-Thinking
- 35.2
Llama 4 Maverick
- 35.0 GPT-5.4 Nano
- 34.7 Claude Sonnet 4
- 33.2 GPT-4o
- 32.1
Pixtral Large
- 31.2
Llama 4 Scout
- 31.1 GPT-5 nano
- 30.0
Qwen2.5-VL-72B
- 29.2
Gemma 3 12B
- 28.8 GPT-4o mini
- 28.8
Mistral Medium 3
- 28.7
Gemma 3 27B
- 28.4 Grok 2 Vision (1212)
- 27.4 GPT-4.1 nano
- 27.2 Claude 3.5 Haiku
- 26.8
Mistral Small 3.1
- 23.0
Gemma 3 4B
- 1
Gemini 3.7 Flash 89.9%
- 2
Gemini 3.5 Flash 86%
- 3
Gemini 3.6 Flash 85.7%
- 4 GPT-5.6 Sol 84.9%
- 5
Gemini 3.1 Pro 81.3%
- 6
Gemini 3 Flash 78%
- 7 GPT-5.5 77.7%
- 8 GPT-5.4 medium 76.3%
- 9
Qwen3.5 397B-A17B 76%
- 10 Kimi K3 76%
- 11
Gemini 3 Pro 75.9%
- 12
Qwen 3.6 Plus 73.3%
- 13
Qwen3.5-Plus 72.6%
- 14 GPT-5.2 medium 67.3%
- 15
Qwen3.6 27B 66.2%
- 16 GPT-5.4 Mini medium 64.2%
- 17
Qwen 3.6 35B-A3B 64.2%
- 18
Qwen 3.6 Flash 63.5%
- 19
Qwen3.5-Flash 61.6%
- 20 Kimi K2.5 60.5%
- 21
Gemini 2.5 Pro Jun 2025 58.1%
- 22
Gemini 3.1 Flash-Lite 56.9%
- 23 GPT-5 medium 51.5%
- 24
Gemini 2.5 Flash Sep 2025 50.6%
- 25 GPT-5 mini medium 49.1%
- 26 GPT-5.1 medium 47.6%
- 27 Grok 4.20 reasoning 46.6%
- 28 o3 46.4%
- 29 Grok 4.3 46.2%
- 30
Qwen3-VL-30B-A3B-Thinking 45.4%
- 31 GLM-4.6V 44.5%
- 32 o4-mini 43.2%
- 33 GPT-4.1 mini 41.7%
- 34 Claude Opus 4 Thinking (low) 41.2%
- 35 GPT-4.1 39.1%
- 36 Claude 3.7 Sonnet 38.7%
- 37 Grok 4 38.7%
- 38
Gemini 2.5 Flash-Lite Sep 2025 38.2%
- 39
Gemini 2.0 Flash Feb 2025 38.1%
- 40 GLM-4.1V-9B-Thinking 37.7%
- 41
Llama 4 Maverick 35.2%
- 42 GPT-5.4 Nano medium 35%
- 43 Claude Sonnet 4 34.7%
- 44 GPT-4o Mar 2025 33.2%
- 45
Pixtral Large 32.1%
- 46
Llama 4 Scout 31.2%
- 47 GPT-5 nano medium 31.1%
- 48
Qwen2.5-VL-72B 30%
- 49
Gemma 3 12B 29.2%
- 50 GPT-4o mini 28.8%
- 51
Mistral Medium 3 28.8%
- 52
Gemma 3 27B 28.7%
- 53 Grok 2 Vision (1212) 28.4%
- 54 GPT-4.1 nano 27.4%
- 55 Claude 3.5 Haiku 27.2%
- 56
Mistral Small 3.1 26.8%
- 57
Gemma 3 4B 23%
Source: Official Visual Reasoning Benchmark (Fab AI) leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 89.9% | Aug 2026 | Official leaderboard | |
| 2 | | 86% | May 2026 | Official leaderboard | |
| 3 | | 85.7% | Jul 2026 | Official leaderboard | |
| 4 | GPT-5.6 Sol | 84.9% | Jul 2026 | Official leaderboard | |
| 5 | | 81.3% | Feb 2026 | Official leaderboard | |
| 6 | | 78% | Dec 2025 | Official leaderboard | |
| 7 | GPT-5.5 | 77.7% | Apr 2026 | Official leaderboard | |
| 8 | GPT-5.4 | 76.3% | medium | Mar 2026 | Official leaderboard |
| 9 | | 76% | Feb 2026 | Official leaderboard | |
| 10 | Kimi K3 | 76% | Jul 2026 | Official leaderboard | |
| 11 | | 75.9% | Nov 2025 | Official leaderboard | |
| 12 | | 73.3% | Mar 2026 | Official leaderboard | |
| 13 | | 72.6% | Feb 2026 | Official leaderboard | |
| 14 | GPT-5.2 | 67.3% | medium | Dec 2025 | Official leaderboard |
| 15 | | 66.2% | Apr 2026 | Official leaderboard | |
| 16 | GPT-5.4 Mini | 64.2% | medium | Mar 2026 | Official leaderboard |
| 17 | | 64.2% | Apr 2026 | Official leaderboard | |
| 18 | | 63.5% | Apr 2026 | Official leaderboard | |
| 19 | | 61.6% | Feb 2026 | Official leaderboard | |
| 20 | Kimi K2.5 | 60.5% | Jan 2026 | Official leaderboard | |
| 21 | | 58.1% | Jun 2025 | Official leaderboard | |
| 22 | | 56.9% | Mar 2026 | Official leaderboard | |
| 23 | GPT-5 | 51.5% | medium | Aug 2025 | Official leaderboard |
| 24 | | 50.6% | Sep 2025 | Official leaderboard | |
| 25 | GPT-5 mini | 49.1% | medium | Aug 2025 | Official leaderboard |
| 26 | GPT-5.1 | 47.6% | medium | Nov 2025 | Official leaderboard |
| 27 | Grok 4.20 | 46.6% | reasoning | Feb 2026 | Official leaderboard |
| 28 | o3 | 46.4% | Apr 2025 | Official leaderboard | |
| 29 | Grok 4.3 | 46.2% | Apr 2026 | Official leaderboard | |
| 30 | | 45.4% | Oct 2025 | Official leaderboard | |
| 31 | GLM-4.6V | 44.5% | Dec 2025 | Official leaderboard | |
| 32 | o4-mini | 43.2% | Apr 2025 | Official leaderboard | |
| 33 | GPT-4.1 mini | 41.7% | Apr 2025 | Official leaderboard | |
| 34 | Claude Opus 4 | 41.2% | Thinking (low) | May 2025 | Official leaderboard |
| 35 | Doubao-1.5 Vision Pro | 39.7% | Official leaderboard | ||
| 36 | GPT-4.1 | 39.1% | Apr 2025 | Official leaderboard | |
| 37 | Claude 3.7 Sonnet | 38.7% | Feb 2025 | Official leaderboard | |
| 38 | Grok 4 | 38.7% | Jul 2025 | Official leaderboard | |
| 39 | | 38.2% | Sep 2025 | Official leaderboard | |
| 40 | | 38.1% | Feb 2025 | Official leaderboard | |
| 41 | GLM-4.1V-9B-Thinking | 37.7% | Jul 2025 | Official leaderboard | |
| 42 | | 35.2% | Apr 2025 | Official leaderboard | |
| 43 | GPT-5.4 Nano | 35% | medium | Mar 2026 | Official leaderboard |
| 44 | Claude Sonnet 4 | 34.7% | May 2025 | Official leaderboard | |
| 45 | GPT-4o Mar 2025 | 33.2% | Mar 2025 | Official leaderboard | |
| 46 | | 32.1% | Nov 2024 | Official leaderboard | |
| 47 | | 31.2% | Apr 2025 | Official leaderboard | |
| 48 | GPT-5 nano | 31.1% | medium | Aug 2025 | Official leaderboard |
| 49 | | 30% | Jan 2025 | Official leaderboard | |
| 50 | | 29.2% | Mar 2025 | Official leaderboard | |
| 51 | GPT-4o mini | 28.8% | Jul 2024 | Official leaderboard | |
| 52 | | 28.8% | May 2025 | Official leaderboard | |
| 53 | | 28.7% | Mar 2025 | Official leaderboard | |
| 54 | Grok 2 Vision (1212) | 28.4% | Dec 2024 | Official leaderboard | |
| 55 | GPT-4.1 nano | 27.4% | Apr 2025 | Official leaderboard | |
| 56 | Claude 3.5 Haiku | 27.2% | Oct 2024 | Official leaderboard | |
| 57 | | 26.8% | Mar 2025 | Official leaderboard | |
| 58 | LFM2.5 VL 1.6B | 26% | Official leaderboard | ||
| 59 | | 23% | Mar 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 89.9% | Aug 2026 | Official leaderboard | |
| 2 | | 86% | May 2026 | Official leaderboard | |
| 3 | | 85.7% | Jul 2026 | Official leaderboard | |
| 4 | GPT-5.6 Sol | 84.9% | Jul 2026 | Official leaderboard | |
| 5 | | 81.3% | Feb 2026 | Official leaderboard | |
| 6 | | 78% | Dec 2025 | Official leaderboard | |
| 7 | GPT-5.5 | 77.7% | Apr 2026 | Official leaderboard | |
| 8 | GPT-5.4 | 76.3% | medium | Mar 2026 | Official leaderboard |
| 9 | | 76% | Feb 2026 | Official leaderboard | |
| 10 | Kimi K3 | 76% | Jul 2026 | Official leaderboard | |
| 11 | | 75.9% | Nov 2025 | Official leaderboard | |
| 12 | | 73.3% | Mar 2026 | Official leaderboard | |
| 13 | | 72.6% | Feb 2026 | Official leaderboard | |
| 14 | GPT-5.2 | 67.3% | medium | Dec 2025 | Official leaderboard |
| 15 | | 66.2% | Apr 2026 | Official leaderboard | |
| 16 | GPT-5.4 Mini | 64.2% | medium | Mar 2026 | Official leaderboard |
| 17 | | 64.2% | Apr 2026 | Official leaderboard | |
| 18 | | 63.5% | Apr 2026 | Official leaderboard | |
| 19 | | 61.6% | Feb 2026 | Official leaderboard | |
| 20 | Kimi K2.5 | 60.5% | Jan 2026 | Official leaderboard | |
| 21 | | 58.1% | Jun 2025 | Official leaderboard | |
| 22 | | 56.9% | Mar 2026 | Official leaderboard | |
| 23 | GPT-5 | 51.5% | medium | Aug 2025 | Official leaderboard |
| 24 | | 50.6% | Sep 2025 | Official leaderboard | |
| 25 | GPT-5 mini | 49.1% | medium | Aug 2025 | Official leaderboard |
| 26 | GPT-5.1 | 47.6% | medium | Nov 2025 | Official leaderboard |
| 27 | Grok 4.20 | 46.6% | reasoning | Feb 2026 | Official leaderboard |
| 28 | o3 | 46.4% | Apr 2025 | Official leaderboard | |
| 29 | Grok 4.3 | 46.2% | Apr 2026 | Official leaderboard | |
| 30 | | 45.4% | Oct 2025 | Official leaderboard | |
| 31 | GLM-4.6V | 44.5% | Dec 2025 | Official leaderboard | |
| 32 | o4-mini | 43.2% | Apr 2025 | Official leaderboard | |
| 33 | GPT-4.1 mini | 41.7% | Apr 2025 | Official leaderboard | |
| 34 | Claude Opus 4 | 41.2% | Thinking (low) | May 2025 | Official leaderboard |
| 35 | GPT-4.1 | 39.1% | Apr 2025 | Official leaderboard | |
| 36 | Claude 3.7 Sonnet | 38.7% | Feb 2025 | Official leaderboard | |
| 37 | Grok 4 | 38.7% | Jul 2025 | Official leaderboard | |
| 38 | | 38.2% | Sep 2025 | Official leaderboard | |
| 39 | | 38.1% | Feb 2025 | Official leaderboard | |
| 40 | GLM-4.1V-9B-Thinking | 37.7% | Jul 2025 | Official leaderboard | |
| 41 | | 35.2% | Apr 2025 | Official leaderboard | |
| 42 | GPT-5.4 Nano | 35% | medium | Mar 2026 | Official leaderboard |
| 43 | Claude Sonnet 4 | 34.7% | May 2025 | Official leaderboard | |
| 44 | GPT-4o Mar 2025 | 33.2% | Mar 2025 | Official leaderboard | |
| 45 | | 32.1% | Nov 2024 | Official leaderboard | |
| 46 | | 31.2% | Apr 2025 | Official leaderboard | |
| 47 | GPT-5 nano | 31.1% | medium | Aug 2025 | Official leaderboard |
| 48 | | 30% | Jan 2025 | Official leaderboard | |
| 49 | | 29.2% | Mar 2025 | Official leaderboard | |
| 50 | GPT-4o mini | 28.8% | Jul 2024 | Official leaderboard | |
| 51 | | 28.8% | May 2025 | Official leaderboard | |
| 52 | | 28.7% | Mar 2025 | Official leaderboard | |
| 53 | Grok 2 Vision (1212) | 28.4% | Dec 2024 | Official leaderboard | |
| 54 | GPT-4.1 nano | 27.4% | Apr 2025 | Official leaderboard | |
| 55 | Claude 3.5 Haiku | 27.2% | Oct 2024 | Official leaderboard | |
| 56 | | 26.8% | Mar 2025 | Official leaderboard | |
| 57 | | 23% | Mar 2025 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as match (figure), match (process) and odd one out. Gemini 3.7 Flash leads 3 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.
match (figure)
57 models we track have a score here.
- 1
Gemini 3.7 Flash 87.4%
- 2
Gemini 3.5 Flash 85.2%
- 3
Gemini 3.6 Flash 84.7%
- 4
Gemini 3.1 Pro ↑1 82%
- 5 GPT-5.6 Sol ↓1 77%
- 6
Gemini 3 Flash 76%
- 7 GPT-5.5 74.3%
- 8
Gemini 3 Pro ↑3 73.8%
- 9
Qwen3.5 397B-A17B 69.9%
- 10 GPT-5.4 ↓2 69.4%
match (process)
57 models we track have a score here.
- 1 GPT-5.6 Sol ↑3 88.9%
- 2
Gemini 3.7 Flash ↓1 88.9%
- 3
Gemini 3 Flash ↑3 77.8%
- 4
Gemini 3.6 Flash ↓1 77.8%
- 5 GPT-5.5 ↑2 75%
- 6
Gemini 3.5 Flash ↓4 75%
- 7 Kimi K3 ↑3 75%
- 8
Gemini 3.1 Pro ↓3 72.2%
- 9
Qwen 3.6 Plus ↑3 66.7%
- 10 GPT-5.4 ↓2 63.9%
odd one out
57 models we track have a score here.
- 1 GPT-5.6 Sol ↑3 88.3%
- 2
Gemini 3.7 Flash ↓1 84.4%
- 3
Gemini 3.5 Flash ↓1 80.5%
- 4 Kimi K3 ↑6 79.2%
- 5
Gemini 3 Pro ↑6 77.9%
- 6
Gemini 3.1 Pro ↓1 77.9%
- 7
Gemini 3.6 Flash ↓4 77.9%
- 8
Qwen 3.6 Plus ↑4 76.6%
- 9 GPT-5.5 ↓2 76.6%
- 10
Qwen3.6 27B ↑5 75.3%
pattern completion (2d)
57 models we track have a score here.
- 1
Gemini 3.1 Pro ↑4 86.3%
- 2
Gemini 3.6 Flash ↑1 83.6%
- 3
Gemini 3.5 Flash ↓1 82.2%
- 4 GPT-5.6 Sol 82.2%
- 5
Gemini 3.7 Flash ↓4 79.5%
- 6
Gemini 3 Flash 75.3%
- 7 GPT-5.5 74%
- 8
Gemini 3 Pro ↑3 72.6%
- 9
Qwen3.5 397B-A17B 71.2%
- 10 Kimi K3 71.2%
pattern completion (linear)
57 models we track have a score here.
- 1
Gemini 3.7 Flash 95.7%
- 2 GPT-5.6 Sol ↑2 92.3%
- 3
Gemini 3.6 Flash 92.3%
- 4
Gemini 3.5 Flash ↓2 91.5%
- 5
Qwen3.5 397B-A17B ↑4 87.2%
- 6 GPT-5.4 ↑2 87.2%
- 7 GPT-5.5 86.3%
- 8
Gemini 3.1 Pro ↓3 84.6%
- 9
Qwen3.5-Plus ↑4 83.8%
- 10
Gemini 3 Flash ↓4 82.1%
reasoning by analogy
57 models we track have a score here.
- 1
Gemini 3.7 Flash 94.4%
- 2
Gemini 3.5 Flash 88.8%
- 3
Gemini 3.6 Flash 87.9%
- 4 GPT-5.6 Sol 86.5%
- 5 Kimi K3 ↑5 83.7%
- 6 GPT-5.4 ↑2 81.4%
- 7
Gemini 3 Flash ↓1 80.9%
- 8
Qwen3.5 397B-A17B ↑1 80%
- 9
Gemini 3.1 Pro ↓4 80%
- 10
Gemini 3 Pro ↑1 79.1%
More effort adds a median 14 points
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official Visual Reasoning Benchmark (Fab AI) leaderboard
More effort adds a median 14 points
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official Visual Reasoning Benchmark (Fab AI) leaderboard
6 models have held number one on the official leaderboard since its release in September 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Visual Reasoning Benchmark (Fab AI) leaderboard
Gemini 3.5 Flash on the official leaderboard passed 85% in May 2026
That was 8 months after its release. By the same age, 16% of the benchmarks from 2025 had passed 85%. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Visual Reasoning Benchmark (Fab AI) 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 Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (59 entries).
- Pass Models scored 57 models we track, plus 2 other entries on its leaderboard.
- 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, which is fine for a retired test. The strongest tested is GPT-5.6 Sol (#14).
- Note Headroom Saturated. The best is 89.9%, past 85% since May 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on fab-ai.org: 65 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.