| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Code Generation | LCB | Speedup7.81 | 75 | |
| Code Generation | LCB v6 | Accuracy57.1 | 54 | |
| Code Generation | LCB v6 | Pass@150.68 | 48 | |
| Code Generation | LCB v5 | Accuracy61.5 | 45 | |
| Code Generation | LCB | Accuracy85.5 | 29 | |
| Code Reasoning | LCB v6 | Accuracy53.6 | 26 | |
| Code Reasoning | LCB | pass@162.46 | 26 | |
| Code Generation | LCB V6 | Accuracy75.9 | 21 | |
| Code Generation | LCB v5 | Score37.63 | 21 | |
| Code Generation | LCB v5 | pass@149.5 | 20 | |
| Code | LCB v6 | Score87.7 | 20 | |
| Code Generation | LCB v6 | Score84.5 | 18 | |
| Code Generation | LCB V2 | Accuracy26.9 | 17 | |
| Code | LCB | AVG@864.3 | 16 | |
| Coding | LCB v6 | Accuracy80.2 | 16 | |
| Code reasoning | LCB v5 | Accuracy33.72 | 16 | |
| Code Generation | LCB | Throughput66.41 | 16 | |
| Function-level synthesis | LCB hard (held-out 25% eval) | Utility Difference (Δπ)33.2 | 13 | |
| Code Generation | LCB v6 (val) | Accuracy (%)78 | 13 | |
| Code | LCB | Speedup8.37 | 12 | |
| Code Generation | LCB | Pass@130.1 | 11 | |
| Code Generation | LCB 2408-2505 | Pass@174.6 | 11 | |
| Code Generation | LCB (test) | Accuracy21.69 | 10 | |
| Reasoning | LCB | Score59.58 | 9 | |
| Code Generation | LCB 128K | Accuracy (%)79.01 | 8 |