Compare / Claude Opus 4.5 vs Grok 4.3
Claude Opus 4.5 vs Grok 4.3
20 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Opus 4.5 | Grok 4.3 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,187 | 1,193 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,459 | 1,428 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,472 | 1,440 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,469 | 1,423 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,503 | 1,452 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,483 | 1,415 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,474 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,467 | 1,424 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 29.1 | 24.9 |
| Artificial Analysisreported | GPQA Diamond% of questions | 86.6% | 90.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 30.1% | 37.2% |
| Artificial Analysisreported | IFBench% of instructions | 58.0% | 83.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.3% | 75.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 47.0% | 37.9% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 89.5% | 97.7% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 45.7% | 33.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 3.0% | 21.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.39 | 0.34 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 70.4 | 57.7 |
| Vending-Bench 2reported | Money after a yearUS dollars | $4,967.06 | $35.26 |
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.