Inference
EN
Anthropic

Claude Sonnet 4 Thinking

Claude Sonnet 4 Thinking is a model from Anthropic. 8 benchmarks count toward its index, in 5 categories.

  • Input: text
RankedEnough results for an index.
IndexAll categories, out of 100.46.9±8.2Median 51.0
CoverageIndex weight that has results.75%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.—Median 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.1 counted50.8—
CodingCode writing and repair.2 counted41.8177 / 209
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.1 counted52.7—
KnowledgeFacts and expert knowledge.2 counted48.1149 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.2 counted46.0103 / 168

Results

8 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH 500Math93.8% ±1.115 / 59
Terminal-Bench 1.0Agentic45.0% ±5.615 / 47
MMLU ProKnowledge83.9% ±0.468 / 135
AIMEMath76.3% ±1.451 / 95
GPQA Diamond · Vals AIKnowledge75.0% ±2.282 / 135
MMMU ProMultimodal74.9% ±1.057 / 92
LiveCodeBenchCoding62.4% ±1.1104 / 142
IOI v1Coding4.6% ±1.146 / 61
MedQAHealthcare92.7% ±0.228 / 95
MGSMMultilingual90.9% ±1.731 / 74
CorpFin v2Finance61.2% ±1.055 / 131
TaxEval v2Finance72.0% ±0.969 / 143
LegalBenchLegal82.1% ±0.575 / 149
MedCodeHealthcare35.0% ±1.978 / 106
MortgageTaxFinance49.9% ±1.084 / 97
MedScribeHealthcare69.4% ±2.296 / 109

Sources

BenchLMCC BY-NC 4.0Vals AINo licence stated

History

5 releases
Index48.0 → 46.9
23 Oct 202612 Nov 2026

Questions

How good is Claude Sonnet 4 Thinking?
Its index is 46.9 out of 100. That is place 170 of 283 ranked models.
What is Claude Sonnet 4 Thinking best at?
Knowledge: 48.1 out of 100, place 149 of 268.

Badge

Claude Sonnet 4 Thinking on Inference360: place 170 of 283, index 46.9
HTML
<a href="https://inference360.ai/models/anthropic-claude-sonnet-4-thinking"><img src="https://inference360.ai/badges/anthropic-claude-sonnet-4-thinking.svg" alt="Claude Sonnet 4 Thinking on Inference360: place 170 of 283, index 46.9" height="20"></a>
Markdown
[![Claude Sonnet 4 Thinking on Inference360: place 170 of 283, index 46.9](https://inference360.ai/badges/anthropic-claude-sonnet-4-thinking.svg)](https://inference360.ai/models/anthropic-claude-sonnet-4-thinking)

Compare with

Gemini 2.5 Pro46.6

More from Anthropic

All models
Claude Fable 5.285.6Claude Opus 5.578.6Claude Sonnet 5.577.2Claude Fable 5.176.2Claude Fable 574.0