invokeable()
← comparisons / claude-sonnet-4-6-vs-claude-sonnet-5

Claude Sonnet 4.6 vs Claude Sonnet 5

Side-by-side from the Invokeable directory — provider-attributed pricing, canonical IDs, draft data flagged. Open as a shareable card →

01Head to head

Claude Sonnet 4.6Claude Sonnet 5
CreatorAnthropicAnthropic
Released2026-02-172026-06-30
Context1M1M
Max output64K128K
Input $/M$3.00/M—
Output $/M$15.00/M—
Knowledge cutoff2025-082026-01
Modalitiestext + image → texttext + image → text
Capabilitiestool-use, extended-thinkingtool-use, extended-thinking
ProvidersAnthropic API, OpenRouterAnthropic API

02Benchmarks · vendor-reported unless verified

Claude Sonnet 4.6Claude Sonnet 5
Artificial Analysis Coding Agent Index (average pass@1)49.4—
BrowseComp (accuracy)—84.7
OSWorld-Verified (success rate)—81.2
SWE-Atlas Codebase QnA (pass@1)70.3—
SWE-Atlas Codebase QnA (score)31.2—
SWE-Atlas Refactoring (score)32.21—
SWE-Atlas Test Writing (score)31.76—
SWE-bench Multilingual—78.3%
SWE-bench Pro—63.2%
SWE-bench Pro (pass@1)14.9—
SWE-bench Verified (resolved)—85.2
Terminal-Bench (success rate)—80.4