Compare / GPT-4.1 Nano vs o3
GPT-4.1 Nano vs o3
15 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | GPT-4.1 Nano | o3 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,321 | 1,426 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,307 | 1,383 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,313 | 1,449 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,300 | 1,403 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,322 | 1,432 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,299 | 1,396 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 83.4% | 86.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 18.9% | 51.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 23.6% | 62.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 33.0% | 63.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 93.8% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 73.0% | 81.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 60.8% | 73.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 11.0% | 77.0% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 6.0% | 49.4% |
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.