Code Generation on LiveCodeBench (LCB) (FUNC)
73.1FUNC ScoreO4-MINI
Evaluation Results
| Method | Links | |
|---|---|---|
| O4-MINIVariant=Generic2026.04 | 73.1 | |
| O4-MINI (Generic Secure Prompt)Variant=Generic Secure2026.04 | 69 | |
| QWQ 32BVariant=Generic2026.04 | 59.4 | |
| SecPI (QWQ 32B)Variant=Our Secure2026.04 | 59.4 | |
| O3Variant=Generic2026.04 | 59.4 | |
| QWQ 32B (Generic Secure Prompt)Variant=Generic Secure2026.04 | 57.6 | |
| SecPI (QWEN 32B-D)Variant=Tuned2026.04 | 55.3 | |
| SecPI (LLAMA 70B-D)Variant=Tuned2026.04 | 55.3 | |
| SecPI (LLAMA 70B-D)Variant=Our Secure2026.04 | 54.7 | |
| QWEN 32B-D (Generic Secure Prompt)Variant=Generic Secure2026.04 | 54.4 | |
| LLAMA 70B-DVariant=Generic2026.04 | 53.8 | |
| QWEN 32B-DVariant=Generic2026.04 | 53.2 | |
| SecPI (QWQ 32B)Variant=Tuned2026.04 | 52.6 | |
| SecPI (QWEN 32B-D)Variant=Our Secure2026.04 | 52.3 | |
| LLAMA 70B-D (Generic Secure Prompt)Variant=Generic Secure2026.04 | 52.3 | |
| O3 (Generic Secure Prompt)Variant=Generic Secure2026.04 | 52.3 | |
| QWEN 14B-DVariant=Generic2026.04 | 52 | |
| SecPI (QWEN 14B-D)Variant=Our Secure2026.04 | 48.5 | |
| QWEN 14B-D (Generic Secure Prompt)Variant=Generic Secure2026.04 | 46.2 | |
| SecPI (QWEN 14B-D)Variant=Tuned2026.04 | 45 | |
| PURPCODEVariant=Generic2026.04 | 26.3 |