Mathematical Reasoning on B-AIME (pass@1)
57.2Pass@1EvoEnv
Evaluation Results
| Method | Links | |
|---|---|---|
| EvoEnvBase Model=Nemotron-Cascade-8B2026.05 | 57.2 | |
| UntrainedBase Model=Nemotron-Cascade-8B2026.05 | 55.8 | |
| RLVEBase Model=Nemotron-Cascade-8B2026.05 | 54.3 | |
| EvoEnvBase Model=Qwen3-4B-Thinking-25072026.05 | 53.6 | |
| UntrainedBase Model=Qwen3-4B-Thinking-25072026.05 | 53.1 | |
| R-ZeroBase Model=Qwen3-4B-Thinking-25072026.05 | 52.9 | |
| R-ZeroBase Model=Nemotron-Cascade-8B2026.05 | 50.1 | |
| RLVEBase Model=Qwen3-4B-Thinking-25072026.05 | 49.6 | |
| DAPOBase Model=Qwen3-4B-Thinking-25072026.05 | 45.8 | |
| DAPOBase Model=Nemotron-Cascade-8B2026.05 | 40.6 | |
| EvoEnvBase Model=Qwen3-4B-Instruct-25072026.05 | 33.6 | |
| RLVEBase Model=Qwen3-4B-Instruct-25072026.05 | 32.8 | |
| UntrainedBase Model=Qwen3-4B-Instruct-25072026.05 | 32 | |
| R-ZeroBase Model=Qwen3-4B-Instruct-25072026.05 | 27 | |
| DAPOBase Model=Qwen3-4B-Instruct-25072026.05 | 22.2 |