LongReason
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Long-context reasoning | LongReason 64K-input 70K context | Accuracy71.25 | 34 | |
| Long-context reasoning | LongReason | Score86.9 | 18 | |
| Multi-choice reasoning | LongReason | Accuracy (32k)84.13 | 17 | |
| Question Answering | LongReason | Acc72.3 | 15 | |
| RL Training | LongReason | Peak Memory (GB)80 | 6 | |
| Reasoning | LongReason (val) | Accuracy (val)79.3 | 4 |
Showing 6 of 6 rows