EnigmaForge - Intuition: leaderboard
EnigmaForge's implicit condition: the model receives only the story, with no stated question, and must discover what problem to solve. Scored by task success on those instances, higher is better.
Metric: Task success (%) on implicit-condition instances, where the question is not stated and must be discovered from the story. Source: robottwo.github.io. Saturation forecast: Around February 2027. 33 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol (Low) | 83.8 | #10 (GPT-6.1 Sol) |
| 2 | GPT-6 Astra (Low) | 80.1 | #6 (GPT-6 Astra) |
| 3 | GPT-6 Sol (Low) | 75 | #25 (GPT-6 Sol) |
| 4 | Gemini 4 Argon (Low) | 72.9 | #11 (Gemini 4 Argon) |
| 5 | GPT-6.1 Terra (Low) | 72.5 | #20 (GPT-6.1 Terra) |
| 6 | Claude Fable 5.2 (Low) | 56.2 | #1 (Claude Fable 5.2) |
| 7 | GPT-5.6 Sol (Low) | 54.2 | #30 (GPT-5.6 Sol) |
| 8 | Gemini 3.7 Flash (Low) | 52.9 | #32 (Gemini 3.7 Flash) |
| 9 | Gemini 3.8 Flash (Low) | 39.3 | #36 (Gemini 3.8 Flash) |
| 10 | Gemini 4 Carbon (Low) | 38.8 | #2 (Gemini 4 Carbon) |
| 11 | GLM-5.3 (Low) | 37.7 | #66 (GLM-5.3) |
| 12 | GPT-5.6 Terra (Low) | 24.2 | #62 (GPT-5.6 Terra) |
| 13 | GPT-6 Luna (Low) | 22.9 | #95 (GPT-6 Luna) |
| 14 | Claude Opus 5.5 (Low) | 22 | #5 (Claude Opus 5.5) |
| 15 | Grok 4.6 (Low) | 21.2 | #45 (Grok 4.6) |
Interactive version: theaggregate.ai/benchmark?slug=enigmaforge-intuition · How It Works · Data refreshed daily, snapshot 2026-11-13.