Game playing on Simple Negotiation
46.2Win RateInstruct Model
Evaluation results
| Method | Links | |
|---|---|---|
| Instruct ModelOpponent=Gemini-2.0-Flash2025.06 | 46.2 | |
| Simple Negotiation SpecialistOpponent=Gemini-2.0-Flash2025.06 | 39.1 | |
| Multi-Game ModelOpponent=Gemini-2.0-Flash2025.06 | 33.2 | |
| TicTacToe SpecialistOpponent=Gemini-2.0-Flash2025.06 | 30.5 | |
| Kuhn Poker SpecialistOpponent=Gemini-2.0-Flash2025.06 | 28.7 | |
| Base ModelOpponent=Gemini-2.0-Flash2025.06 | 15.6 | |
| Random PolicyOpponent=Gemini-2.0-Flash2025.06 | 8.2 |