Math on MMath (Accuracy)
38.33Accuracy (MMath)standard cascade
Evaluation Results
| Method | Links | |
|---|---|---|
| standard cascadeModel=GPT-4o-mini2026.06 | 38.33 | |
| context-aware cascadeModel=GPT-4o-mini2026.06 | 38.33 | |
| end-to-endModel=GPT-4o-mini2026.06 | 36 | |
| context-aware cascadeModel=Llama-3.1-8B-Instruct2026.06 | 16.2 | |
| end-to-endModel=Llama-3.1-8B-Instruct2026.06 | 13.8 | |
| standard cascadeModel=Llama-3.1-8B-Instruct2026.06 | 13.5 | |
| context-aware cascadeModel=Mistral-7B-Instruct-v0.32026.06 | 5.7 | |
| standard cascadeModel=Mistral-7B-Instruct-v0.32026.06 | 5.4 | |
| end-to-endModel=Mistral-7B-Instruct-v0.32026.06 | 4.1 |