Conceptual Reasoning Index - Argument Evaluation (LMCA): leaderboard
Metric: Chance-Corrected Score (0-100). Source: conceptualreasoning.ai. 136 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 (Max) | 63.3 |
| 2 | Claude Fable 5 (Max) | 60.34 |
| 3 | GPT-5.6 Sol (Max) | 58.43 |
| 4 | Claude Opus 4.8 (Max) | 57.51 |
| 5 | Claude Opus 4.6 (Max) | 55.76 |
| 6 | GPT-5.5 (xHigh) | 54.27 |
| 7 | GPT-5.5 Pro (xHigh) | 53.94 |
| 8 | Gemini 3.1 Pro (Preview) (High) | 53.82 |
| 9 | Kimi K3 (Max) | 52.74 |
| 10 | GPT-5.6 Terra (Max) | 52.23 |
| 11 | Claude Opus 4.7 (Max) | 52.2 |
| 12 | GPT-5.4 (xHigh) | 51.97 |
| 13 | Gemini 3.7 Flash (High) | 50.38 |
| 14 | Muse Spark 1.1 (High) | 49.91 |
| 15 | Claude Sonnet 5 (Max) | 49.31 |
Interactive version: theaggregate.ai/benchmark?slug=conceptual-reasoning-index-argument-evaluation-lmca · How It Works · Data refreshed daily, snapshot 2026-09-05.