2048 Game on 2048 4x4 grid
4,662.7ScoreDoubao-Seed-1.8
Evaluation Results
| Method | Links | |
|---|---|---|
| Doubao-Seed-1.82026.02 | 4,662.7 | |
| Base Model + GLAD + MC-CriticBackbone=Qwen3-4B-Instruct2026.02 | 4,503.8 | |
| GPT-52026.02 | 4,040 | |
| Base Model + GLADBackbone=Qwen3-4B-Instruct2026.02 | 3,335.3 | |
| UITARS-1.52026.02 | 2,616 | |
| Doubao-Seed-1.62026.02 | 1,877.3 | |
| Qwen3-30B-A3B-Instruct2026.02 | 838.7 | |
| Qwen3-4B-Instruct (Base Model)2026.02 | 721.3 | |
| Qwen3-235B-A22B-Instruct2026.02 | 634.7 | |
| Claude-4.5-Sonnet2026.02 | 166.7 |