StakeBench: leaderboard
Existing financial NLP benchmarks often rely on labels supplied by outside observers, measuring how language is perceived rather than what speakers have committed to in the market. We introduce StakeBench, an evaluation framework for language understanding grounded in market commitment. StakeBench links 560,876 comments from 2,261 resolved markets to verified position, action, and market-odds reco
Metric: Agg (self-reported). Source: benchmarklist.com. Saturation forecast: Around June 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 20.4 | #31 |
| 2 | Qwen 3 32B | 17.6 | #443 |
| 3 | Gemini 2.5 Flash | 17.5 | #245 |
| 4 | Qwen 3 14B | 17.3 | #537 |
| 5 | Gemma 2 9B | 15.2 | #958 |
| 6 | Qwen 3 30B A3B | 14 | #493 |
| 7 | Claude Haiku 4.5 | 13.8 | #280 |
| 8 | Qwen 3 8B | 10.3 | #674 |
| 9 | Llama 3.2 3B | 9.2 | #1396 |
| 10 | Mistral 7B | 4.5 | #1397 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=stakebench · How It Works · Data refreshed daily, snapshot 2026-10-20.