Compare / Claude Opus 4.1 vs GPT-5.4 Mini
Claude Opus 4.1 vs GPT-5.4 Mini
16 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 4.1 | GPT-5.4 Mini |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,435 | 1,447 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,447 | 1,460 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,445 | 1,402 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,482 | 1,480 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,458 | 1,434 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,450 | 1,447 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,446 | 1,424 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 22.8 | 24.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 80.9% | 87.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 12.5% | 28.1% |
| Artificial Analysisreported | IFBench% of instructions | 55.4% | 73.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 76.0% | 77.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 34.3% | 52.3% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 71.4% | 83.3% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 92.4% | 100.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 11.8% | 5.5% |
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.