Compare / Claude Haiku 4.5 vs GPT-4.1 Nano
Claude Haiku 4.5 vs GPT-4.1 Nano
14 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 Haiku 4.5 | GPT-4.1 Nano |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,420 | 1,321 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,387 | 1,307 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,452 | 1,313 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,415 | 1,300 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,414 | 1,322 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,399 | 1,299 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 95.3% | 83.4% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 54.4% | 18.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 53.6% | 23.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 68.7% | 33.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 62.5% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 86.5% | 73.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 78.7% | 60.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 83.5% | 11.0% |
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.