Othello
8×8 Othello at 20 empties. Edax solves every legal move exactly by final disc difference; the answer is a cell. See Methodology.
Quality is the mean oracle score of the moves played, × 100. The best move scores 100; an illegal move or no move scores 0. Protocol boardbench-0.2.
Quality vs cost
7 of 7 models
- Anthropic
- OpenAI
- DeepSeek
- Z.ai
- Meta
- Pareto frontier: nothing plotted is both cheaper and better
- Anthropic
- OpenAI
- DeepSeek
- Z.ai
- Meta
Horizontal axis is logarithmic.
Table
| Model | Quality | Legality | Cost | n |
|---|---|---|---|---|
| LunaHigh | 55.1 | 100.0 | <$0.01 | 3 |
| SolHigh | 55.1 | 100.0 | $0.10 | 3 |
| Muse Spark 1.3High | 55.1 | 100.0 | $0.11 | 3 |
| GLM-5.3 FlashHigh | 37.8 | 100.0 | $0.02 | 3 |
| DeepSeek V4.1 FlashHigh | 34.5 | 100.0 | $0.04 | 3 |
| Haiku 5.5High | 2.7 | 66.7 | <$0.01 | 3 |
| Opus 5.5High | 0.4 | 100.0 | $0.04 | 3 |
Headline is the mean move score × 100. n is answers: puzzles × samples.