Compare / Claude Sonnet 5 vs Grok 4.5
Claude Sonnet 5 vs Grok 4.5
28 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 Sonnet 5 | Grok 4.5 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 54.5% | 56.2% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.01 | -0.01 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.09 | -0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.07 | 0.04 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,476 | 1,464 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,196 | 1,202 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,452 | 1,465 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,465 | 1,465 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,435 | 1,452 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,512 | 1,488 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,467 | 1,462 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,462 | 1,465 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,452 | 1,459 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 71.5 | 72.4 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 38.2 | 38.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 91.1% | 93.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 41.3% | 42.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.0% | 79.3% |
| Artificial Analysisreported | SciCode% of problems | 54.3% | 55.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 80.5% | 81.6% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 14.1% | 10.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 37.3% | 42.1% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 33.7% | 48.3% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 3.0% | 21.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.38 | 0.32 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 62.7 | 60.0 |
| Vending-Bench 2reported | Money after a yearUS dollars | $6,377.70 | $3,887.43 |
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.