Visual Cipher Bench - Knowledge: leaderboard
Visual Cipher Bench v1 (2611.16861v1): 40 image-based puzzles, ten each of spatial, logic, rule and knowledge types. Harbor v0.24.0; GPT models use Codex CLI v0.163.0 and other models Claude Code v2.1.296; 60-minute network-disabled runs, 2 CPUs/4 GB/no GPU, Python/NumPy/Pillow tools and read-only puzzle images. Three independent runs per exact model/effort. Accuracy is final-answer exact match after fixed normalization; invalid submissions/timeouts score zero. Each category has 30 attempts; overall has 120. Literal none effort is retained without inferring disabled reasoning. Mean solved-count column is not another benchmark. This board reports knowledge exact-answer accuracy in percent; higher is better.
Metric: Exact final-answer accuracy (%). Source: arxiv.org. Saturation forecast: Around 2029. 8 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-6 Astra | Max | 13.3 | #6 |
| 2 | GPT-6.1 Sol | Max | 13.3 | #10 |
| 3 | Claude Opus 5.5 | xHigh | 10 | #5 |
| 4 | DeepSeek V4.1 Flash | Max | 0 | #62 |
| 5 | GPT-6 Luna | 0 | #99 | |
| 6 | Claude Sonnet 5.5 | Max | 0 | #19 |
| 7 | Kimi K3.1 Pro | Max | 0 | #29 |
| 8 | Qwen 3.8 Max | xHigh | 0 | #59 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=visual-cipher-bench-knowledge · How It Works · Data refreshed daily, snapshot 2026-11-16.