Exact-solution programming on LiveCodeBench v5
96.8Easy Exact Match RateQwen3-8B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen3-8BBackbone=Qwen3-8B2026.06 | 96.8 | 62.9 | 28.8 | 56.1 | |
| Rank-uniformBackbone=Qwen3-8B2026.06 | 96.8 | 61.5 | 31.1 | 56.7 | |
| RiVERBackbone=Qwen3-8B2026.06 | 96.7 | 67.9 | 30.2 | 58.3 | |
| RiVERBackbone=GLM-Z1-9B-04142026.06 | 96.7 | 64.7 | 24.3 | 54.7 | |
| Raw-GRPOBackbone=Qwen3-8B2026.06 | 95.9 | 64.1 | 29.7 | 56.7 | |
| RS-GRPOBackbone=Qwen3-8B2026.06 | 95.9 | 59.6 | 31.1 | 55.9 | |
| Raw-BinaryBackbone=Qwen3-8B2026.06 | 95.9 | 62.2 | 27.5 | 55.1 | |
| Instance-NormBackbone=Qwen3-8B2026.06 | 95.9 | 62.2 | 28.8 | 55.7 | |
| Raw-GRPOBackbone=GLM-Z1-9B-04142026.06 | 95.1 | 66.7 | 20.7 | 53.3 | |
| Instance-NormBackbone=GLM-Z1-9B-04142026.06 | 95.1 | 64.7 | 22.5 | 53.5 | |
| RS-GRPOBackbone=GLM-Z1-9B-04142026.06 | 94.3 | 61.5 | 17.6 | 50.1 | |
| Raw-BinaryBackbone=GLM-Z1-9B-04142026.06 | 93.5 | 62.2 | 24.8 | 53.3 | |
| GLM-Z1-9B-0414Backbone=GLM-Z1-9B-04142026.06 | 93.4 | 57.7 | 18.6 | 49.2 | |
| Rank-uniformBackbone=GLM-Z1-9B-04142026.06 | 92.7 | 62.8 | 20.7 | 51.5 |