anthropic Vision Reasoning

Claude Opus 4 pricing & benchmarks

Released May 22, 2025 · served by 1 provider

Input / 1M tokens$15.00
Output / 1M tokens$75.00
Context window200K
Cached input$1.50
1M in + 300K out$37.50

What Claude Opus 4 is

Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows. It sets new benchmarks in...

Providers and prices for Claude Opus 4

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
Googlecheapest $15.00 $75.00 200K 100.0%

Claude Opus 4 benchmark results

Design Arena head-to-head results, as reported through the OpenRouter model API.

CategoryArenaRankEloWin rate
svg models #39 1177 57.7%
gamedev models #41 1227 59.9%
3d models #47 1196 57.7%
uicomponent models #55 1190 59.2%
codecategories models #58 1191 55.6%
website models #59 1190 54.6%
dataviz models #60 1175 57.9%

FAQ

How much does the Claude Opus 4 API cost?

$15.00 per 1M input tokens and $75.00 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $37.50.

Is Claude Opus 4 free?

No. It is a paid model starting at $15.00 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for Claude Opus 4?

Google at $15.00 input / $75.00 output per 1M tokens.

Can I self-host Claude Opus 4?

No. Claude Opus 4 is closed-weight and only available through hosted APIs.