Misguided Attention - jugs_4_liters: leaderboard

Each task is one independently graded modified logic puzzle with native expected-answer criteria. Zero-shot; source's per-task mean over available valid evaluations, not a completion rate over nominal attempts. Native num_evaluations is the only supported denominator. Models.js and query_config files supply registry settings; no per-row execution timestamp or request log match is claimed. Evaluation code weights criterion satisfaction and clips each run to0–1 before summarizing; do not apply new clipping to published data. Current evaluator names Gemini2.5Flash at temperature0/max1000, while older README names Llama3.3; exact judge version for each historic run is unlogged and must stay unknown. Public prompts are available, including reversibly scrambled source files. Metric: Mean native expected-behavior score (0–1). Source: https://cpldcpu.github.io/MisguidedAttention/

Metric: Mean native expected-behavior score (0–1). Source: cpldcpu.github.io. Saturation forecast: Estimated already saturated. 53 models tracked. Updated 17 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-4.1 Mini1#793
2GPT-4.11#505
3DeepSeek R11#487
4Gemini 2.5 Pro (Preview 03-25)1#239
5Grok 31#606
6DeepSeek V3 (0324)1#722
7DeepSeek R1 05281#431
8Grok 3 Mini Beta1#620
9O3 Mini1#547
10Grok 3 Mini1#546
11O4 MiniHigh1#367
12Qwen 3 235B A22BThinking1#631
13Gemini 2.5 Pro (Preview 05-06)1#271
14Mistral Medium 31#1097
15Qwen 3 30B A3BThinking1#1155

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=misguided-attention-jugs-4-liters · How It Works · Data refreshed daily, snapshot 2026-11-17.