Mathematical Reasoning on AIME 2024 (avg@16)
85.8AIME 2024 Avg ScoreASTER-4B
Evaluation results
| Method | Links | |
|---|---|---|
| ASTER-4BModel Category=Agentic Reasoning Models, Inference Budget=90K2026.02 | 85.8 | |
| Qwen3-235B-A22B-ThinkingModel Category=Text-Only Reasoning Models2026.02 | 85.7 | |
| OpenReasoning-Nemotron-7BModel Category=Text-Only Reasoning Models2026.02 | 84.7 | |
| ASTER-4BModel Category=Agentic Reasoning Models2026.02 | 82.3 | |
| POLARIS-4B-PreviewModel Category=Text-Only Reasoning Models2026.02 | 81.2 | |
| rStar2-Agent-14BModel Category=Agentic Reasoning Models2026.02 | 80.6 | |
| OpenAI o3-mini (medium)Model Category=Text-Only Reasoning Models2026.02 | 79.6 | |
| Qwen3-4B-Thinking-2507Model Category=Text-Only Reasoning Models2026.02 | 76.7 | |
| DemyAgent-4BModel Category=Agentic Reasoning Models, Inference Budget=30K2026.02 | 72.6 | |
| ReTool-32BModel Category=Agentic Reasoning Models2026.02 | 72.5 | |
| ASTER-1.7BModel Category=Agentic Reasoning Models2026.02 | 64.6 | |
| ASTER-4B-SFTModel Category=Agentic Reasoning Models2026.02 | 62.5 | |
| Qwen3-1.7B-ThinkingModel Category=Text-Only Reasoning Models2026.02 | 47.5 | |
| ASTER-1.7B-SFTModel Category=Agentic Reasoning Models2026.02 | 19.4 |