Claude Haiku 4.5 scores 18.2 on the Respan Index (#105), ahead of DeepSeek R1 at 17.9 (#107). Claude Haiku 4.5 leads on 4 of the 7 benchmarks both report. Claude Haiku 4.5 has the larger context window (200K tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 1414 |
| 1373 |
| BALROG (BALROG) | 31.2% | 34.9% |
| SWE-bench Verified | 73.3% | 49.2% |
Higher is better. Each score links to where it was published.