Strategic Proposer-Solver Game on Arm QC (question committed) ten games (test)
7Total Score (G)gpt-5.5
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| gpt-5.52026.07 | 7 | 0.368 | 0.148 | 0.207 | |
| gpt-5.42026.07 | 6.5 | 0.363 | 0.144 | 0.198 | |
| opus-4.82026.07 | 4.4 | 0.324 | 0.187 | 0.14 | |
| qwen3.7-max2026.07 | 2.8 | 0.319 | 0.197 | 0.09 | |
| sonnet-4.62026.07 | 2.1 | 0.276 | 0.198 | 0.088 | |
| haiku-4.52026.07 | 1.1 | 0.246 | 0.197 | 0.097 | |
| gpt-5.4-mini2026.07 | 0.4 | 0.246 | 0.219 | 0.062 | |
| gpt-5.4-nano2026.07 | -0.5 | 0.175 | 0.174 | 0.054 | |
| kimi-k2.72026.07 | -4.8 | 0.171 | 0.253 | 0.03 | |
| deepseek-v4-pro2026.07 | -5.6 | 0.278 | 0.335 | 0.023 | |
| gpt-4o-mini2026.07 | -13.3 | 0.104 | 0.388 | 0.011 |