Inference
EN
Anthropic

Claude Sonnet 4.5

Claude Sonnet 4.5 is a standard model from Anthropic. 21 benchmarks count toward its index, in 5 categories.

  • Closed weights
  • Standard
  • Input: text · image · file
  • Released 2025-09
RankedEnough results for an index.
IndexAll categories, out of 100.47.9±5.5Median 51.0
CoverageIndex weight that has results.80%
SpeedOutput tokens per second.45/sMedian 67/s
First tokenSeconds to the first output token.1.0 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$3.00batch $1.50Median $0.60
Output / 1MUS dollars per 1M output tokens.$15.00batch $7.50 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.1439±3 #80Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
anthropic/claude-sonnet-4.5
Max outputMaximum output tokens per request.
64K
Knowledge cutoffLast date of training data.
31 Jan 2025
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.30
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
6 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
AnthropicDeprecatedMaker$3.00$15.00200K
DatabricksCloud$3.00$15.00200K
DigitalOceanCloud$3.00$15.00200K
OpenRouterRouter$3.00$15.001M
SAP AI CoreCloud$3.00$15.00200K
Snowflake CortexCloud$3.00$15.00200K

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.6 counted49.3133 / 218
CodingCode writing and repair.1 counted53.9—
ReasoningLogic problems and puzzles.5 counted45.3143 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.2 counted49.4138 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.7 counted46.697 / 168

Results

21 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH level 5Math97.7% ±0.45 / 88
SWE-bench Verified · SWE-bench/experimentsCoding71.4%11 / 65
Software Engineering Benchmark VerifiedCoding77.2%36 / 114
τ²-bench AirlineAgentic71.0%8 / 22
τ²-bench TelecomAgentic91.4%7 / 18
GPQA diamond · Epoch AIKnowledge80.3% ±2.988 / 217
Graduate-Level Google-Proof Q&AKnowledge83.4%59 / 141
American Invitational Mathematics Examination 2025Math87.0%23 / 53
Gert Labs Composite Game BenchmarkAgentic48.5%24 / 52
OTIS Mock AIME 2024-2025Math74.4% ±6.393 / 200
FrontierMath-2025-02-28-PrivateMath13.5% ±2.034 / 69
τ²-bench RetailAgentic79.3%13 / 23
FrontierMath-Tier-4-2025-07-01-PrivateMath3.1% ±2.933 / 55
Mystery Game PuzzlesReasoning17.0% ±3.851 / 80
τ²-bench BankingAgentic25.3%19 / 28
ARC-AGI-2 (semi-private)Reasoning3.8%89 / 131
SWE-Bench verified · Epoch AICoding71.3% ±2.123 / 32
JobBenchAgentic27.7%26 / 36
Terminal-Bench 2.0 · BenchLMAgentic50.0%30 / 41
ARC-AGI-1 (semi-private)Reasoning25.5%99 / 126
SimpleQA VerifiedKnowledge27.2% ±1.572 / 91
FrontierMath-Tiers-1-3-v2-PrivateMath23.9% ±2.575 / 90
OSWorld-VerifiedAgentic61.4%32 / 37
Chess PuzzlesReasoning10.0% ±3.079 / 90
FrontierMath-Tier-4-v2-PrivateMath2.4% ±2.465 / 72
VITA-BenchAgentic17.0%12 / 13
EBR-benchReasoning2.4%30 / 31
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
MedScribeHealthcare84.5% ±1.934 / 109
CorpFin v2Finance60.8% ±1.063 / 131
FrontierMath-2025-02-28-PublicMath0.0%33 / 64
SWE-bench Multilingual · SWE-bench/experimentsCoding67.0%8 / 14
MedCodeHealthcare40.6% ±2.063 / 106
ARC-AGI-2 (public eval)Reasoning3.8%78 / 122
ARC-AGI-1 (public eval)Reasoning35.4%93 / 116

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYSierra ResearchMITSWE-bench teamNo licence statedARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Index48.7 → 47.9
23 Oct 202612 Nov 2026
Output / 1M$15.00 → $15.00
23 Oct 202612 Nov 2026

Questions

How much does Claude Sonnet 4.5 cost?
$3.00 per 1M input tokens, $15.00 per 1M output tokens.
How good is Claude Sonnet 4.5?
Its index is 47.9 out of 100. That is place 162 of 283 ranked models.
What is Claude Sonnet 4.5 best at?
Knowledge: 49.4 out of 100, place 138 of 268.
How fast is Claude Sonnet 4.5?
45 output tokens per second, 1.0 s to the first token.
What is the context window of Claude Sonnet 4.5?
1M tokens.
Is Claude Sonnet 4.5 open source?
No. Its weights are not published.
When was Claude Sonnet 4.5 released?
September 2025.

Badge

Claude Sonnet 4.5 on Inference360: place 162 of 283, index 47.9
HTML
<a href="https://inference360.ai/models/claude-sonnet-4-5"><img src="https://inference360.ai/badges/claude-sonnet-4-5.svg" alt="Claude Sonnet 4.5 on Inference360: place 162 of 283, index 47.9" height="20"></a>
Markdown
[![Claude Sonnet 4.5 on Inference360: place 162 of 283, index 47.9](https://inference360.ai/badges/claude-sonnet-4-5.svg)](https://inference360.ai/models/claude-sonnet-4-5)

Compare with

Gemma 4 31B47.9Claude Opus 4.147.5Claude Haiku 4.5 Thinking43.5Claude Sonnet 4.5 Thinking53.8

Lower price

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

More from Anthropic

All models
Claude Sonnet 4.5 Thinking53.8Claude Haiku 4.5 Thinking43.5Claude Haiku 4.543.2Claude Opus 4.147.5Claude 4.1 Opus Thinking47.0