EnigmaForge: leaderboard

Metric: Task success (%) over 600 procedurally generated story instances, all items (missing or invalid answers count as failures). Source: robottwo.github.io. Saturation forecast: Around January 2027. 22 models tracked.

Top models

#ModelScore
1GPT-6 (Low)80.3
2GPT-6 Sol (Low)74.2
3Gemini 3.7 Flash (Low)68.7
4GPT-5.6 Sol (Low)68.5
5Gemini 3.8 Flash (Low)59
6GLM-5.3 (Low)53.2
7GPT-5.6 Terra (Low)50.7
8GPT-6 Luna (Low)42.3
9GPT-5.6 Luna (Low)41.7
10Claude Opus 5.5 (Low)39.8
11Grok 4.6 (Low)37
12Qwen 3.8 27B (Low)33.7

Interactive version: theaggregate.ai/benchmark?slug=enigmaforge · How It Works · Data refreshed daily, snapshot 2026-09-29.