Claude Opus 5

Name: Claude Opus 5
Author: Anthropic

Multimodal

Anthropic

Claude Opus 5 is Anthropic's flagship general-access model, built for efficient agentic coding, knowledge work, scientific research, computer use, and problem-solving. Anthropic reports state-of-the-art results on Frontier-Bench v0.1 and GDPval-AA, substantial gains over Opus 4.8 across life-sciences evaluations, and support for configurable effort levels.

Key Specifications

Parameters

Context

1.0M

Release Date

July 24, 2026

Average Score

55.2%

API Documentation Research Paper Results Blog

Timeline

Key dates in the model's history

Announcement

July 24, 2026

Last Update

August 27, 2026

Technical Specifications

Parameters

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)

$5.00

Output (per 1M tokens)

$25.00

Max Input Tokens

1.0M

Max Output Tokens

128.0K

Supported Features

Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Other Tests

Specialized benchmarks

ARC-AGI-3

Novel problem-solving. • Self-reported

30.2%

AutomationBench

Business workflows. • Self-reported

26.0%

BioMysteryBench

Human-solved subset. • Self-reported

90.1%

BrowseComp

Agentic search. • Self-reported

90.8%

DeepSWE 1.1

Agentic coding. • Self-reported

68.8%

Frontier-Bench v0.1

Agentic terminal coding. Anthropic's internal run on the mini-SWE-agent harness and a GKE backend; mean reward over five attempts per task. Opus 4.8 was the fallback for safety-classifier refusals. • Self-reported

43.3%

FrontierCode

FrontierCode v1.1 Main subset. • Self-reported

53.4%

FrontierCode 1.1

FrontierCode 1.1 current leaderboard; mergeability score at medium effort: 53.4%. • Self-reported

53.4%

HealthBench Professional

HealthBench Professional. • Self-reported

59.8%

Humanity's Last Exam

Multidisciplinary reasoning with tools. • Self-reported

64.7%

Legal Agent Benchmark

Legal Agent Benchmark held-out set. • Self-reported

11.7%

OSWorld 2.0

Computer use. • Self-reported

70.6%

License & Metadata

License

proprietary

Announcement Date

July 24, 2026

Last Updated

August 27, 2026

Compare Claude Opus 5

All comparisons

vs Claude Sonnet 5 vs Claude Fable 5 vs GPT-5.6 Luna vs GPT-5.6 Terra vs Grok 4.6 vs GPT-5.6 Sol vs Qwen3.8 Flash Next vs Muse Spark 1.2

Similar Models

All Models

Claude Sonnet 5

Anthropic

Released:Jun 2026

Price:$2.00/1M tokens

Claude 3.7 Sonnet

Anthropic

Best score:0.8 (GPQA)

Released:Feb 2025

Price:$3.00/1M tokens

Claude 3 Sonnet

Anthropic

Best score:0.9 (ARC)

Released:Feb 2024

Price:$3.00/1M tokens

Claude Sonnet 4.6

Anthropic

Best score:0.9 (GPQA)

Released:Feb 2026

Price:$3.00/1M tokens

Claude Opus 4.8

Anthropic

Best score:0.9 (GPQA)

Released:May 2026

Price:$5.00/1M tokens

Claude Opus 4.1

Anthropic

Best score:0.8 (TAU)

Released:Aug 2025

Price:$15.00/1M tokens

Claude 3 Haiku

Anthropic

Best score:0.9 (ARC)

Released:Mar 2024

Price:$0.25/1M tokens

Claude Haiku 4.5

Anthropic

Best score:0.8 (TAU)

Released:Oct 2025

Price:$1.00/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.