Inference
EN
Anthropic

Claude 3.5 Sonnet

Claude 3.5 Sonnet is a standard model from Anthropic. 12 benchmarks count toward its index, in 4 categories.

  • Closed weights
  • Standard
  • Input: text
  • Released 2024-06
RankedEnough results for an index.
IndexAll categories, out of 100.32.2±9.2Median 51.0
CoverageIndex weight that has results.55%
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.$3.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$15.00Median $2.55
ContextMaximum tokens in one request.200KMedian 262K
EloLMArena rating and rank.1298±3 #256Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.None counted——
CodingCode writing and repair.3 counted34.0193 / 209
ReasoningLogic problems and puzzles.None counted——
MultimodalTasks with images and text.1 counted44.7—
KnowledgeFacts and expert knowledge.2 counted37.4212 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.6 counted25.2151 / 168

Results

12 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
SWE-bench Lite · SWE-bench/experimentsCoding40.8%4 / 15
SWE-bench Lite · swebench.comCoding32.5%4 / 15
MATH level 5Math51.7% ±1.247 / 88
SWE-bench Verified · SWE-bench/experimentsCoding53.0%38 / 65
GPQA diamond · Epoch AIKnowledge54.0% ±2.8138 / 217
SWE-bench Verified · swebench.comCoding33.6%31 / 46
MMLU ProKnowledge78.4% ±0.4103 / 135
MMMU ProMultimodal68.8% ±1.172 / 92
Graduate-Level Google-Proof Q&AKnowledge59.4%111 / 141
GPQA Diamond · Vals AIKnowledge59.3% ±2.5110 / 135
FrontierMath-Tier-4-2025-07-01-PrivateMath0.0%45 / 55
LiveCodeBenchCoding49.6% ±1.1117 / 142
OTIS Mock AIME 2024-2025Math6.5% ±2.2165 / 200
MATH 500Math72.4% ±3.949 / 59
FrontierMath-2025-02-28-PrivateMath1.0% ±0.659 / 69
Software Engineering Benchmark VerifiedCoding49.0%100 / 114
AIMEMath10.0% ±0.988 / 95
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
MGSMMultilingual92.6% ±1.615 / 74
SWE-bench (full test split)Coding21.5%3 / 7
MortgageTaxFinance64.1% ±1.848 / 97
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
TaxEval v2Finance70.2% ±0.993 / 143
MedQAHealthcare83.2% ±0.764 / 95
CorpFin v2Finance53.6% ±1.098 / 131
SWE-bench Multimodal · SWE-bench/experimentsMultimodal8.8%6 / 7
SWE-bench Multimodal · swebench.comMultimodal8.8%8 / 8

Sources

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

History

5 releases
Index33.3 → 32.2
23 Oct 202612 Nov 2026
Output / 1M$15.00 → $15.00
23 Oct 202612 Nov 2026

Questions

How much does Claude 3.5 Sonnet cost?
$3.00 per 1M input tokens, $15.00 per 1M output tokens.
How good is Claude 3.5 Sonnet?
Its index is 32.2 out of 100. That is place 246 of 283 ranked models.
What is Claude 3.5 Sonnet best at?
Knowledge: 37.4 out of 100, place 212 of 268.
What is the context window of Claude 3.5 Sonnet?
200K tokens.
Is Claude 3.5 Sonnet open source?
No. Its weights are not published.
When was Claude 3.5 Sonnet released?
June 2024.

Badge

Claude 3.5 Sonnet on Inference360: place 246 of 283, index 32.2
HTML
<a href="https://inference360.ai/models/claude-3-5-sonnet"><img src="https://inference360.ai/badges/claude-3-5-sonnet.svg" alt="Claude 3.5 Sonnet on Inference360: place 246 of 283, index 32.2" height="20"></a>
Markdown
[![Claude 3.5 Sonnet on Inference360: place 246 of 283, index 32.2](https://inference360.ai/badges/claude-3-5-sonnet.svg)](https://inference360.ai/models/claude-3-5-sonnet)

Compare with

Magistral Small31.4Claude 3.5 Haiku24.1Claude 3 Haiku22.0

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from Anthropic

All models
Claude 3 Opus23.3Claude 3 Haiku22.0Claude 3.5 Haiku24.1Claude 3.7 Sonnet Thinking41.4Claude 3.7 Sonnet37.7