Claude 4 Sonnet
Anthropic · 2 thinking configurations
Intelligence18.9
Input / 1M$3.00
Output / 1M$15.00
Context1M
Tariff: OpenRouter · Source: openrouter · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost2 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Achieving state-of-the-art performance on SWE-bench (72.7%),...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
18.9 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
77.7 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
10.7 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
36.3 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
64.6 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
54.7 %MMMU-ProMultimodal understanding of academic problems that include images.
61.8 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
22.7 %Specifications from the catalog
Inputimage, text, file
Outputtext
Tool callingYes
Structured outputNo
SpeedNot available
First tokenNot available
Maximum output64K
Catalog listing date2025-05-22