Ideation Arena - Feasibility: leaderboard
Metric: Elo (Bradley-Terry over expert pairwise votes). Source: arxiv.org. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.1 | 1180.9 |
| 2 | Claude Sonnet 4.5 | 1163.6 |
| 3 | Claude Opus 4.5 | 1160.2 |
| 4 | Mistral Large 3 | 1102.2 |
| 5 | DeepSeek V3.2 | 1073.9 |
| 6 | Kimi K2 (Thinking) | 1060.7 |
| 7 | O3 | 1050 |
| 8 | GLM-4.7 | 1049.7 |
| 9 | Gemini 3 Flash (Preview) | 1038.5 |
| 10 | Grok 4 Fast | 1025.1 |
| 11 | Gemini 3 Pro (Preview) | 1020.1 |
| 12 | Qwen 3 Max | 993.2 |
| 13 | Grok 4 | 983.1 |
| 14 | GPT-4o | 925 |
Interactive version: theaggregate.ai/benchmark?slug=ideation-arena-feasibility · How It Works · Data refreshed daily, snapshot 2026-09-19.