Pseudo-Deliberation in Language Models — leaderboard
Metric: Fast Thinking Macro F1 (self-reported). Source: benchmarklist.com. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3 8B Instruct | 47.7 |
| 2 | Gemini 3 Flash (Preview) | 47.7 |
| 3 | GPT-4o | 45.3 |
| 4 | Qwen 3 8B | 44.2 |
Interactive version: theaggregate.ai/benchmark?slug=pseudo-deliberation-in-language-models · How the rankings work · Data refreshed daily, snapshot 2026-07-22.