// BENCHMARK

AutoLogi benchmark

AI model leaderboard for the AutoLogi benchmark. Compare how large language models score on AutoLogi, see the full ranking, and understand what this AI benchmark measures. Qwen3-14B currently leads with 89.2. Open-ended logic-puzzle benchmark — automatically constructed deduction problems with verifiable solutions.

Leaderboard

# Model Organization Score Variant Source
#1 Qwen3-14B Qwen 89.2 thinking official ↗
#2 Qwen3-8B Qwen 89.1 thinking official ↗
#3 Qwen3-235B-A22B Qwen 89 thinking official ↗
#4 Qwen3-30B-A3B Qwen 88.7 thinking official ↗
#5 Qwen3-4B Qwen 87.9 thinking official ↗
#6 Qwen3-32B Qwen 87.3 thinking official ↗
#7 Qwen3-235B-A22B Qwen 83.3 non-thinking official ↗
#8 Qwen3-1.7B Qwen 83.2 thinking official ↗
#9 Qwen3-14B Qwen 82 non-thinking official ↗
#10 Qwen3-30B-A3B Qwen 81.5 non-thinking official ↗
#11 Qwen3-32B Qwen 78.5 non-thinking official ↗
#12 Qwen3-8B Qwen 76.5 non-thinking official ↗
#13 Qwen3-4B Qwen 76.3 non-thinking official ↗
#14 Qwen3-0.6B Qwen 61.6 thinking official ↗
#15 Qwen3-1.7B Qwen 59.8 non-thinking official ↗
#16 Qwen3-0.6B Qwen 37.4 non-thinking official ↗

Frequently asked questions about AutoLogi

What is the AutoLogi benchmark?

Open-ended logic-puzzle benchmark — automatically constructed deduction problems with verifiable solutions.

How is the AutoLogi benchmark scored?

AutoLogi is scored using the accuracy metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.

Which AI model scores highest on AutoLogi?

As of the latest reported scores on GenAIList, Qwen3-14B achieves the highest result on AutoLogi with a score of 89.2.

Is a higher AutoLogi score better?

Yes. On AutoLogi a higher score indicates better performance, so models near the top of the leaderboard are the strongest.