Compare / Claude Opus 4.8 vs Claude Opus 5.5
Claude Opus 4.8 vs Claude Opus 5.5
21 results from 6 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 4.8 | Claude Opus 5.5 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 48.9% | 73.5% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.04 | 0.16 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.15 | 0.19 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.10 | 0.14 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,490 | 1,499 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,467 | 1,521 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,525 | 1,556 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,491 | 1,516 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,480 | 1,509 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,474 | 1,520 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 41.8 | 57.6 |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 48.7% | 61.4% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.7% | 84.7% |
| Artificial Analysisreported | SciCode% of problems | 54.4% | 66.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 21.7% | 59.6% |
| GDP.pdfreported | Professional document tasks% of rubric | 24.0% | 30.6% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 53.0% | 72.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 6.0% | 2.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.40 | 0.43 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 65.9 | 70.0 |
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.