Leaderboard
PolyBench
LLM forecasting and trading on Polymarket binary markets. Each point-in-time snapshot gives the model the order book and live news, and its trades are executed against the simulated order book.
PolyBench arXiv v1 (2026-04-03)
Confidence-Weighted Return (CWR) %, higher is better. 7 models.
Updated 3 Apr 2026
- MiMo-V2-Flash17.6
- Gemini-3-Flash6.2
- Trinity-Large-9.2
- deepseek-v3.2-11.4
- Grok-4.1-Fast-12.1
- GPT-OSS 120B-17.8
- MiniMax-M2.5-25.2
PolyBench arXiv v1 (2026-04-03)
Confidence-Weighted Return (CWR) %, higher is better. 7 models.
Updated 3 Apr 2026
- 1MiMo-V2-Flash17.6
- 2Gemini-3-Flash6.2
- 3Trinity-Large-9.2
- 4deepseek-v3.2-11.4
- 5Grok-4.1-Fast-12.1
- 6GPT-OSS 120B-17.8
- 7MiniMax-M2.5-25.2
| 1 | MiMo-V2-Flash | Xiaomi | 17.6 | base lot $10, 2026-02-06 to 2026-02-12 | Paper Table 5 Reported by the authors, 3 Apr 2026 |
|---|---|---|---|---|---|
| 2 | Gemini-3-Flash | 6.2 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 | |
| 3 | Trinity-Large | Not stated | -9.2 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 |
| 4 | deepseek-v3.2 | DeepSeek | -11.4 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 |
| 5 | Grok-4.1-Fast | xAI | -12.1 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 |
| 6 | GPT-OSS 120B | OpenAI | -17.8 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 |
| 7 | MiniMax-M2.5 | MiniMax | -25.2 | base lot $10 | Paper Table 5 Reported by the authors, 3 Apr 2026 |