Compare / Claude Haiku 4.5 vs Claude Opus 5
Claude Haiku 4.5 vs Claude Opus 5
19 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Haiku 4.5 | Claude Opus 5 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,419 | 1,490 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,418 | 1,488 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,420 | 1,486 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,387 | 1,472 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,452 | 1,543 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,415 | 1,497 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,414 | 1,491 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,399 | 1,482 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 43.9 | 78.0 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 16.9 | 50.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 67.2% | 93.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 10.4% | 54.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 74.3% | 82.0% |
| Artificial Analysisreported | SciCode% of problems | 42.2% | 56.4% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 44.2% | 89.1% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 49.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 9.3% | 44.7% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 13.2% | 59.9% |
| Vending-Bench 2reported | Money after a yearUS dollars | $458.89 | $11,181.87 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.