Compare / claude-3-haiku-20240307 vs GPT-5.4
claude-3-haiku-20240307 vs GPT-5.4
9 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | claude-3-haiku-20240307 | GPT-5.4 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,248 | 1,482 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,216 | 1,443 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,253 | 1,515 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,245 | 1,471 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,262 | 1,475 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,247 | 1,465 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 15.0% | 30.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.34 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 31.9 | 68.8 |
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.