Head-to-head comparison across 3benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
Claude Opus 4.5
80
Claude Sonnet 4.6
85
Verified leaderboard positions: Claude Opus 4.5 #7 · Claude Sonnet 4.6 unranked
Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. Claude Opus 4.5 only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.
Agentic
+2.8 difference
Coding
+0.5 difference
Knowledge
+7.5 difference
Claude Opus 4.5
Claude Sonnet 4.6
$null / $null
$3 / $15
46 t/s
44 t/s
1.01s
1.48s
200K
200K
Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. Claude Opus 4.5 only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.
Claude Sonnet 4.6 is clearly ahead on the provisional aggregate, 85 to 80. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Claude Sonnet 4.6's sharpest advantage is in knowledge, where it averages 73.7 against 66.2. The single biggest benchmark swing on the page is SuperGPQA, 70.6% to 95%.
Claude Sonnet 4.6 is ahead on BenchLM's provisional leaderboard, 85 to 80. The biggest single separator in this matchup is SuperGPQA, where the scores are 70.6% and 95%.
Claude Sonnet 4.6 has the edge for knowledge tasks in this comparison, averaging 73.7 versus 66.2. Inside this category, SuperGPQA is the benchmark that creates the most daylight between them.
Claude Sonnet 4.6 has the edge for coding in this comparison, averaging 66.4 versus 65.9. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Claude Sonnet 4.6 has the edge for agentic tasks in this comparison, averaging 65.3 versus 62.5. Inside this category, OSWorld-Verified is the benchmark that creates the most daylight between them.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.