Strategic Negotiation on Frontier Model Negotiation Scenarios Seller Role GFT/NGFT
100IR ComplianceGem.-2.5-Pro
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Gem.-2.5-ProRole=Seller2026.04 | 100 | 72.6 | 91.7 | 2.06 | 94 | 99.3 | |
| o3Role=Seller2026.04 | 99.95 | 77 | 95.6 | 3.04 | 96 | 100 | |
| GPT-4.1Role=Seller2026.04 | 99.9 | 57.7 | 95.2 | 2.07 | 96.2 | 97.7 | |
| Gem.-2.5-FlashRole=Seller2026.04 | 99.9 | 57.6 | 94.9 | 2.14 | 93.4 | 98.1 | |
| DeepSeek-V3Role=Seller2026.04 | 99.3 | 49.6 | 92.1 | 1.97 | 90 | 99.4 |