Skip to main content
ModelScale

Claude Sonnet 4.5

Anthropic · Proprietary · rank 96 · bench-align-v5

Compare this model Model a workload
Canonical idclaude-sonnet-4-5
Overall score53.96
Context window200K tokens
Release date2025-09-01
Access typeProprietary
Blended $/1M$6.0075% input / 25% output

Capability shape

Seven axes from the ranking source. A missing axis is drawn as a gap.

Capability evidence

Agentic33.7
Coding59.5
Knowledge72.8
Reasoning19.1
Multimodal & GroundedUnavailable
Instruction FollowingUnavailable
Math34.8

Runtime service evidence

Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.

Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.

Evidence key: ObservedUnavailable

Runtime measurements
MeasurementValueObservedLast goodEvidence
Time to first tokenUnavailableUnavailableUnavailable
ThroughputUnavailableUnavailableUnavailable

Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.

Endpoint and price matrix

Every published price component, including cache reads and writes.

Price components
ComponentUSDEvidence
Input / 1M tokens$3.00
Output / 1M tokens$15.00
Cache read / 1M tokensUnavailable
Cache write / 1M tokensUnavailable
Blended / 1M (75% input / 25% output)$6.00
Self-hosted listingAnthropic's Claude Sonnet 4.6 announcement says pricing remains the same as Sonnet 4.5 at $3 input / $15 output per million tokens.The rates above are a hosted price matched from another provider, not a first-party list price.

Workload-aware monthly cost example

10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.

Modelled monthly cost$16.90
Modelled tokens2.82M
Open the simulatorChange this workload

Benchmark record

12 matched benchmark rows with their published value, unit, and provenance.

Knowledge

Knowledge benchmarks
BenchmarkValueTasksDifficultyProvenance
GPQA
Graduate-Level Google-Proof Q&A
83.4448 questionsGraduate levelGPQA: A Graduate-Level Google-Proof Q&A Benchmark

Coding

Coding benchmarks
BenchmarkValueTasksDifficultyProvenance
SWE-bench Verified
Software Engineering Benchmark Verified
77.2500 verified issuesProfessional software engineeringSWE-bench: Can Language Models Resolve Real-World GitHub Issues?

Mathematics

Mathematics benchmarks
BenchmarkValueTasksDifficultyProvenance
AIME 2025
American Invitational Mathematics Examination 2025
8715 problemsHigh school olympiad levelAmerican Invitational Mathematics Examination
FrontierMath v2 (Tiers 1-3)
FrontierMath v2 Tiers 1-3
13.495295 private advanced mathematics problemsFrom olympiad-plus to early research mathematicsFrontierMath v2 benchmark hub
FrontierMath v2 (Tier 4)
FrontierMath v2 Tier 4
4.16743 private extreme-difficulty mathematics problemsResearch-level mathematics requiring hours or days of expert workFrontierMath Tier 4 v2 leaderboard

Reasoning

Reasoning benchmarks
BenchmarkValueTasksDifficultyProvenance
ARC-AGI-2
Abstraction and Reasoning Corpus for AGI v2
13.6Visual pattern completion and abstract reasoningExpert-level — hardest public reasoning benchmarkARC-AGI-2: A Harder General Intelligence Benchmark

Agentic

Agentic benchmarks
BenchmarkValueTasksDifficultyProvenance
Terminal-Bench 2.0
Terminal-Bench 2.0
50Terminal-based software tasksProfessional software engineeringTerminal-Bench 2.0
Gert Labs
Gert Labs Composite Game Benchmark
48.51Novel game environmentsAgentic coding and decision-makingGert Labs rankings
OSWorld-Verified
OSWorld-Verified
61.4369 real-world computer tasks (361 when eight Google Drive tasks are excluded)Multi-step desktop and cross-application workflowsOSWorld
JobBench
JobBench
27.7130 tasks across 35 occupationsProfessional multi-source workflowsJobBench: Aligning Agent Work With Human Will
VITA-Bench
VITA-Bench
17.0Interactive consumer-service agent tasksLong-horizon real-world workflowsVitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications

Multimodal & Grounded

Multimodal & Grounded benchmarks
BenchmarkValueTasksDifficultyProvenance
Design Arena Website
Design Arena Website Elo
1202Website generation comparisonsDesign and website generationOpenRouter Grok 4.3 benchmarks

Lifecycle and limitations log

Lifecycle events the source associates with this model, plus what this profile does not claim.

No lifecycle event references this model.

That is not evidence the model has no lifecycle plan — only that this source published none.

What this profile does not claimValues are reproduced exactly as their sources published them, in the units those sources declared; none are converted, interpolated, or averaged across providers. Any field marked unavailable was attempted and not returned — the attempt timestamp is in each badge. Last attempted fetch for this model’s score: 2026-09-17 17:17 UTC.