Misguided Attention - no_paradox_expected_hanging: 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

#ModelScoreOverall rank
1GLM-Z1-32B0.6667
2DeepSeek R1 Distill Llama 70B0.3333#1669
3GPT-4o0#754
4Llama 3.3 70B Instruct0#1387
5GPT-4.1 Mini0#793
6GPT-4.10#505
7GPT-4.1 Nano0#2043
8Gemini 2.0 Flash0#723
9DeepSeek R10#487
10Llama 4 Maverick0#1118
11DeepSeek V30#734
12Claude Sonnet 40#411
13Gemini 2.5 Pro (Preview 03-25)0#239
14Claude 3.7 Sonnet0#476
15Gemma 3 27B0#1671

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-no-paradox-expected-hanging · How It Works · Data refreshed daily, snapshot 2026-11-17.