Inference
EN
Anthropic

Claude 3.7 Sonnet

Claude 3.7 Sonnet is a standard model from Anthropic. 21 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Standard
  • Input: text
  • Released 2025-02
RankedEnough results for an index.
IndexAll categories, out of 100.37.7±4.0Median 51.0
CoverageIndex weight that has results.95%
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.200KMedian 262K
EloLMArena rating and rank.1299±4 #253Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.1 counted40.7—
CodingCode writing and repair.3 counted42.1176 / 209
ReasoningLogic problems and puzzles.2 counted39.8165 / 227
MultimodalTasks with images and text.2 counted40.376 / 84
KnowledgeFacts and expert knowledge.5 counted40.0193 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.2 counted38.145 / 63
MathMath problems.6 counted30.0143 / 168

Results

21 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
SWE-bench Lite · swebench.comCoding48.0%2 / 15
SWE-bench Lite · SWE-bench/experimentsCoding48.0%2 / 15
MATH level 5Math88.1% ±0.818 / 88
Instruction-Following EvalInstruction90.8%25 / 70
SWE-bench Verified · SWE-bench/experimentsCoding62.4%27 / 65
τ²-bench AirlineAgentic64.2%10 / 22
GPQA diamond · Epoch AIKnowledge76.8% ±3.099 / 217
SWE-bench Verified · swebench.comCoding52.8%22 / 46
Artificial Analysis Omniscience AccuracyKnowledge28.2%84 / 164
Massive Multi-discipline Multimodal UnderstandingMultimodal71.8%17 / 28
Artificial Analysis IFBenchInstruction44.0%90 / 146
OTIS Mock AIME 2024-2025Math50.0% ±7.5124 / 200
MMLU ProKnowledge80.7% ±0.485 / 135
τ²-Bench Tool-Agent-User EvaluationAgentic50.0%96 / 149
MMMLUKnowledge83.2%10 / 15
Graduate-Level Google-Proof Q&AKnowledge68.0%98 / 141
Critical Physics TasksReasoning0.0%150 / 214
Artificial Analysis Long Context ReasoningReasoning51.7%153 / 217
Artificial Analysis GPQA DiamondKnowledge65.6%142 / 199
MATH 500Math76.8% ±1.943 / 59
Software Engineering Benchmark VerifiedCoding62.3%84 / 114
MMMU ProMultimodal71.5% ±1.168 / 92
GPQA Diamond · Vals AIKnowledge67.4% ±2.4100 / 135
FrontierMath-2025-02-28-PrivateMath3.3% ±1.052 / 69
LiveCodeBenchCoding56.7% ±1.1111 / 142
Artificial Analysis MMMU-ProMultimodal60.1%92 / 113
AIMEMath22.3% ±1.479 / 95
τ²-bench RetailAgentic72.1%20 / 23
Artificial Analysis Humanity's Last ExamKnowledge4.2%194 / 222
SWE-Bench verified · Epoch AICoding61.0% ±2.228 / 32
τ²-bench TelecomAgentic49.0%16 / 18
MATH-500 Problem SetMath82.2%25 / 27
American Invitational Mathematics Examination 2024Math23.3%26 / 27
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
MortgageTaxFinance68.7% ±1.811 / 97
MGSMMultilingual92.4% ±1.618 / 74
SWE-bench (full test split)Coding33.8%2 / 7
TaxEval v2Finance72.4% ±0.962 / 143
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
LegalBenchLegal80.0% ±0.591 / 149
SWE-bench Multimodal · swebench.comMultimodal30.9%5 / 8
SWE-bench Multimodal · SWE-bench/experimentsMultimodal30.9%5 / 7

Sources

BenchLMCC BY-NC 4.0Vals AINo licence statedEpoch AICC BYSierra ResearchMITSWE-bench teamNo licence stated

History

5 releases
Index38.8 → 37.7
23 Oct 202612 Nov 2026

Questions

How good is Claude 3.7 Sonnet?
Its index is 37.7 out of 100. That is place 224 of 283 ranked models.
What is Claude 3.7 Sonnet best at?
Instruction: 38.1 out of 100, place 45 of 63.
What is the context window of Claude 3.7 Sonnet?
200K tokens.
Is Claude 3.7 Sonnet open source?
No. Its weights are not published.
When was Claude 3.7 Sonnet released?
February 2025.

Badge

Claude 3.7 Sonnet on Inference360: place 224 of 283, index 37.7
HTML
<a href="https://inference360.ai/models/claude-3-7-sonnet"><img src="https://inference360.ai/badges/claude-3-7-sonnet.svg" alt="Claude 3.7 Sonnet on Inference360: place 224 of 283, index 37.7" height="20"></a>
Markdown
[![Claude 3.7 Sonnet on Inference360: place 224 of 283, index 37.7](https://inference360.ai/badges/claude-3-7-sonnet.svg)](https://inference360.ai/models/claude-3-7-sonnet)

Compare with

Mistral Small 437.9Granite 4.2 8B37.6DeepSeek V3.137.9DeepSeek V3 032437.1Claude 3.7 Sonnet Thinking41.4

More from Anthropic

All models
Claude 3.7 Sonnet Thinking41.4Claude 4 Opus Thinking47.0Claude 4 Sonnet39.5Claude 4 OpusUnrankedClaude Sonnet 443.0