Compare / Claude Haiku 4.5 vs Claude Opus 4.8
Claude Haiku 4.5 vs Claude Opus 4.8
25 results from 5 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 4.8 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,419 | 1,486 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,418 | 1,469 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,420 | 1,490 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,387 | 1,467 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,452 | 1,525 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,415 | 1,491 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,414 | 1,480 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,399 | 1,474 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 43.9 | 74.3 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 16.9 | 41.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 67.2% | 92.0% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 10.4% | 48.7% |
| Artificial Analysisreported | IFBench% of instructions | 54.3% | 62.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 74.3% | 77.7% |
| Artificial Analysisreported | SciCode% of problems | 42.2% | 54.4% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 44.2% | 84.6% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 21.7% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 27.3% | 58.3% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 9.3% | 34.2% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 54.7% | 94.4% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 13.2% | 53.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 6.0% | 6.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.39 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 52.2 | 65.9 |
| Vending-Bench 2reported | Money after a yearUS dollars | $458.89 | $5,787.43 |
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.