Formal Theorem Proving on Synthetic 20
14.5Proof Length
GPT-5.3-Codex
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.04 | 14.5 | |
2026.04 | 15.4 | |
2026.04 | 16.9 | |
2026.04 | 17.3 | |
2026.04 | 19.2 | |
2026.04 | 21.1 |
GPT-5.3-Codex
| Method | Links | |
|---|---|---|
2026.04 | 14.5 | |
2026.04 | 15.4 | |
2026.04 | 16.9 | |
2026.04 | 17.3 | |
2026.04 | 19.2 | |
2026.04 | 21.1 |