StakeBench - Commitment Detection: leaderboard
Metric: Accuracy (%) on G1: does a comment come from a user holding a position in the market (yes or no, chance 50), one-shot prompt, Polymarket and Manifold comments of resolved markets, macro-average over 18 topic and platform splits, greedy decoding; higher is better. Source: arxiv.org. Saturation forecast: Around May 2028. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 32B (Non-reasoning) | 74.7 |
| 2 | Qwen 3 14B (Non-reasoning) | 74 |
| 3 | DeepSeek R1 Distill Llama 8B | 72.1 |
| 4 | Gemma 2 9B | 71.8 |
| 5 | Gemini 2.5 Flash | 71.1 |
| 6 | Claude Haiku 4.5 | 69.7 |
| 7 | Qwen 3 8B (Non-reasoning) | 69.2 |
| 8 | GPT-5.5 | 68.9 |
| 9 | Qwen 3 30B A3B (Non-reasoning) | 67 |
| 10 | Llama 3.2 3B Instruct | 56.7 |
| 11 | Mistral 7B Instruct (v0.3) | 53.4 |
Interactive version: theaggregate.ai/benchmark?slug=stakebench-commitment-detection · How It Works · Data refreshed daily, snapshot 2026-10-07.