Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d67, 15 pendingSources1,014

Claude Sonnet 4.5

AnthropicReleased Sep 29, 2025Current

Score basis

Claude Sonnet 4.5 is capable in long context; and behind the leaders in multimodal tasks, coding, instruction following, factuality, reasoning, and math. Too few results yet to rate agentic tasks, safety, or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Capable1 capability
  1. Long Context−16.9%2 of 3
Limited6 capabilities
  1. Multimodal−30.9%4 of 6
  2. Coding−31.9%8 of 10
  3. Instruction Following−34.6%2 of 3
  4. Factuality−34.8%4 of 4
  5. Reasoning−35.5%5 of 6
  6. Math−54.4%2 of 5
Not rated3 capabilities

Too few results yet to rate Agentic, Safety or Multilingual.

Price

$3.00input$15.00outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Costs more than 88% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

197results on127benchmarks

  • 44 independently verified
  • 35 aggregator
  • 26 vendor-reported
  • 92 cross-referenced

From 40 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

43 papers reference Claude Sonnet 4.5

Claude Sonnet 4.5 benchmark results

197 results on 127 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

CapableFull long context ranking

16.9% behind the leader2 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Multimodal

LimitedFull multimodal ranking

30.9% behind the leader4 of 6 ranked benchmarks measured

Show 2 more multimodal resultsHide 2 multimodal results

Coding

LimitedFull coding ranking

31.9% behind the leader8 of 10 ranked benchmarks measured

Show 15 more coding resultsHide 15 coding results

Instruction Following

LimitedFull instruction following ranking

34.6% behind the leader2 of 3 ranked benchmarks measured

Show 2 more instruction following resultsHide 2 instruction following results

Factuality

LimitedFull factuality ranking

34.8% behind the leader4 of 4 ranked benchmarks measured

Show 3 more factuality resultsHide 3 factuality results

Reasoning

LimitedFull reasoning ranking

35.5% behind the leader5 of 6 ranked benchmarks measured

Show 12 more reasoning resultsHide 12 reasoning results

Math

LimitedFull math ranking

54.4% behind the leader2 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

Agentic

Not enough dataFull agentic ranking

0 of 7 ranked benchmarks measured

Show 6 more agentic resultsHide 6 agentic results

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 118 more resultsHide 118 results

Claude Sonnet 4.5: common questions

Who makes Claude Sonnet 4.5?

Claude Sonnet 4.5 is made by Anthropic.

When was Claude Sonnet 4.5 released?

Claude Sonnet 4.5 was released on Sep 29, 2025, according to Artificial Analysis.

What is Claude Sonnet 4.5 good at?

Claude Sonnet 4.5 is capable in long context; and behind the leaders in multimodal tasks, coding, instruction following, factuality, reasoning, and math. Too few results yet to rate agentic tasks, safety, or multilingual tasks.

How much does Claude Sonnet 4.5 cost?

Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens, according to Artificial Analysis. We track its price at 2 providers. At a mix of three input tokens to one output token, it costs more than 88% of the 330 priced models we track.

How many benchmarks has Claude Sonnet 4.5 been tested on?

We track 197 results for Claude Sonnet 4.5 on 127 benchmarks from 40 sources, 44 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does Claude Sonnet 4.5 support?

OpenRouter lists tool calling, structured outputs, and reasoning for Claude Sonnet 4.5.

About this record

Where Claude Sonnet 4.5's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Aug 27, 2026

Where the results come from

Verification: 197 scores · 44 independently verified · 35 aggregator-attributed · 92 vendor cross-reference · 26 vendor-reported. How these tiers are assigned

From 40 sources on 15 sites. Hugging Face supplies 92 of them; the 44 independently verified results come from 11 sites. Bars are coloured by trust tier.

  • huggingface.co92
  • artificialanalysis.ai38
  • www-cdn.anthropic.com14
  • api.llm-stats.com9
  • arcprize.org9
  • storage.googleapis.com6
  • epoch.ai5
  • swebench.com5
  • raw.githubusercontent.com4
  • anthropic.com3
  • datasets-server.huggingface.co3
  • labs.scale.com3
  • matharena.ai3
  • lmarena.ai2
  • simple-bench.com1

Also known as

Anthropic Claude Sonnet 4.5claude 4.5 sonnetclaude 4.5 sonnet (20250929)claude 4.5 sonnet (high reasoning)Claude 4.5 Sonnet (high)claude 4.5 sonnet (non-reasoning)claude 4.5 sonnet (reasoning)claude code + sonnet 4.5claude sonnet 4.5 (32k thinking)claude sonnet 4.5 (59k thinking)claude sonnet 4.5 (no thinking)claude sonnet 4.5 (thinking 16k)claude sonnet 4.5 (thinking 1k)claude sonnet 4.5 (thinking 32k)claude sonnet 4.5 (thinking 8k)claude sonnet 4.5 (thinking)and 12 more