Compare / Claude Opus 5 vs gpt-oss-120b
Claude Opus 5 vs gpt-oss-120b
18 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 5 | gpt-oss-120b |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 65.8% | 4.4% |
| Arena (formerly LMArena)reported | OverallArena rating | 1,488 | 1,385 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,486 | 1,351 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,472 | 1,277 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,543 | 1,357 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,497 | 1,323 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,491 | 1,352 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,482 | 1,309 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 78.0 | 30.4 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 50.8 | 11.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 93.7% | 78.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 54.9% | 19.6% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.0% | 52.0% |
| Artificial Analysisreported | SciCode% of problems | 56.4% | 34.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 89.1% | 26.2% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 49.0% | 0.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 44.7% | 12.8% |
| Vending-Bench 2reported | Money after a yearUS dollars | $11,181.87 | $-21.53 |
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.