Head-to-head comparison across 3benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
Claude Sonnet 4.5
68
Claude Sonnet 4.6
85
Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. Claude Sonnet 4.5 only becomes the better choice if coding is the priority.
Agentic
+10.0 difference
Coding
+10.8 difference
Knowledge
+9.7 difference
Claude Sonnet 4.5
Claude Sonnet 4.6
$3 / $15
$3 / $15
N/A
44 t/s
N/A
1.48s
200K
200K
Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. Claude Sonnet 4.5 only becomes the better choice if coding is the priority.
Claude Sonnet 4.6 is clearly ahead on the provisional aggregate, 85 to 68. 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 agentic, where it averages 65.3 against 55.3. The single biggest benchmark swing on the page is OSWorld-Verified, 61.4% to 72.5%. Claude Sonnet 4.5 does hit back in coding, so the answer changes if that is the part of the workload you care about most.
Claude Sonnet 4.6 is ahead on BenchLM's provisional leaderboard, 85 to 68. The biggest single separator in this matchup is OSWorld-Verified, where the scores are 61.4% and 72.5%.
Claude Sonnet 4.5 has the edge for knowledge tasks in this comparison, averaging 83.4 versus 73.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Claude Sonnet 4.5 has the edge for coding in this comparison, averaging 77.2 versus 66.4. 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 55.3. 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.