Claude Opus 4.5

Anthropic's most capable model, for complex analysis, research and long-context work.

200,000context window
64,000max output
4ways to run it
Anthropicmade by

Capabilities

toolsjson_schemastreamingprompt_cachingbatchcitations

text · image · pdf → text

Where you can run it

Illustrative figures for layout only — not verified pricing. The live catalogue is not yet sourced.

ProviderModel IDAPIPrice / M tokensRegions
Anthropicclaude-opus-4-5-20251101anthropic$5.00 in · $25.00 outglobal
Amazon Bedrockanthropic.claude-opus-4-5-v1:0bedrock_runtime$5.00 in · $25.00 outus-east-1, eu-central-1
OpenRouteranthropic/claude-opus-4.5openai_compatible$5.25 in · $26.25 outglobal
Tedoclaude-opus-4.5anthropic€5.50 in · €27.00 outeu

The same model. Different identifiers, different prices, different regions.

When to use it

Long documents, careful analysis, and anything where being wrong is expensive. It is the model to reach for when the task is hard rather than large.

When not to

High-volume classification or extraction. You will pay five times Sonnet for accuracy you did not need.

Common questions

Is it the same model on every provider?

Yes — the weights are identical. What differs is the identifier you call, the price, the regions, and occasionally the maximum output.