Anthropic · Claude Opus 4

Claude Opus 4.1

RetiredStale · Sep 10, 2026

Early Opus-tier model in the Claude 4 line, now legacy. Uses a 200k-token context window with extended thinking and vision at the standard Opus price of $5/$25 per million tokens.

profile normalized against the 105-model field

Context window· 200K of 10M2%

Max output· 32K of 384K8%

Output speed—

Affordability· $75.00 / Mtok out58%

Capability breadth· 4 of 1136%

Capability switches · 4 of 11

Reasoning mode

Tool / function use

Streaming

JSON mode

Structured outputs

Prompt caching

Fine-tuning

Web search

Code execution

Vision input

Audio input

This model has been retired

Anthropic shut this model down. It is no longer available, and the specifications and pricing below are kept for historical reference only.

Specifications

Every value carries a primary source and a verification date.

Capacity

Context window

200K

Max output

32K

Pricing

Input $/Mtok

$15 / MTok USD per million tokens

Cached input $/Mtok

$1.50 / MTok cached input USD per million tokens

Output $/Mtok

$75 / MTok USD per million tokens

Capabilities

Input modalities

Text, image

Output modalities

Text

Reasoning mode

Extended thinking: Yes

Tool / function use

Yes

Prompt caching

Yes

Vision input

Yes

API

API model ID

claude-opus-4-1

Batch API

Yes

General

Knowledge cutoff

January 2025

Release date

August 5, 2025

Benchmarks

Sourced evaluation scores, each verified against its primary source.

GPQA Diamond

Claude Opus 4.1 reaches GPQA Diamond: 80.9%.

80.9 %Verified

SWE-bench Verified

On SWE-bench Verified, Claude Opus 4.1 scores 74.5%.

74.5 %Verified

Humanity's Last Exam (no tools)

Humanity's Last Exam (no tools): 11.2% for Claude Opus 4.1.

11.2 %Verified

AIME 2025

AIME 2025: 78.0% with extended thinking.

78 %Verified

MMLU-Pro

MMLU-Pro: 83.5% for Claude Opus 4.1.

83.5 %Verified

Loading…