Claude Opus 5

Anthropic · Released 24 Jul 2026 · claude-opus-5

Compare this model

ProprietaryCodingAgentsReasoning

$5.00

Input price / 1M tokens

#52 of 61, cheapest first

$25.00

Output price / 1M tokens

#52 of 61, cheapest first

Context window

#29 of 71

Output tokens / second

#51 of 61

52.3%

Terminal-Bench 4.0

#6 of 10

48%

FrontierCode 1.1

#3 of 6

26.9%

AutomationBench

#7 of 8

63.6%

Humanity’s Last Exam (with tools)

#6 of 16

Summary

Claude Opus 5 is a proprietary model from Anthropic, released on 24 Jul 2026.

At $5.00 input and $25.00 output per million tokens, it is #52 of 61 on input price, cheapest first.

Its 1M-token context window ranks #29 of 71.

Measured output speed is 55 tokens per second, #51 of 61.

Legacy Opus model, superseded by Claude Opus 5.5. Anthropic described it as coming close to Claude Fable 5's intelligence at half the price, designed for everyday use.

Benchmarks · 10 reported

Benchmark	Score	Bar	Rank	Source
Terminal-Bench 4.0Anthropic setup; public leaderboard (5 trials/task, Claude Code harness) reports 51.8%	52.3%		#6 of 10	Anthropic Opus 5.5 launch post (comparison column)Self-reported
FrontierCode 1.1comparison column in a later post	48%		#3 of 6	Anthropic Opus 5.5 launch postSelf-reported
CursorBench 4.0comparison column in a later post	46.6%		–	Anthropic Opus 5.5 launch postSelf-reported
GDPval-AA v2.1comparison column in a later post	1,708	Elo-style	–	Anthropic Opus 5.5 launch postSelf-reported
AutomationBenchpass rate from Zapier's public leaderboard	26.9%		#7 of 8	Zapier public leaderboard, as cited in Anthropic Opus 5.5 launch post
Humanity’s Last Exam (with tools)with tools	63.6%		#6 of 16	Anthropic Opus 5.5 launch postSelf-reported
Humanity’s Last Exam (no tools)no tools	56.6%		#3 of 23	Anthropic Fable 5.1 / Mythos 5.1 launch post (comparison column)Self-reported
Terminal-Bench-Science 0.1Anthropic setup; public leaderboard (3 trials/task, Claude Code harness) reports 30.0%	29%		–	Anthropic Opus 5.5 launch postSelf-reported
OSWorld 2.1partial-credit score	74%		–	Anthropic Opus 5.5 launch postSelf-reported
Chartographywith tools	83.4%		–	Anthropic Opus 5.5 launch postSelf-reported

Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.

Similar models

Models with the most benchmarks in common with Claude Opus 5, and the closest scores.

Compare 4 side by side

Terminal-Bench 4.0

#6 of 10

Long jobs in a real terminal

FrontierCode 1.1

#3 of 6

Patches good enough to merge

AutomationBench

#7 of 8

End-to-end business workflows

Humanity’s Last Exam (with tools)

#6 of 16

Expert questions, with search and code

Compare Claude Opus 5 side by side

Pre-filled with the two closest models. Swap any of them.

Open comparison

Sources

price, cache prices, batch, fast mode, tokenizer: platform.claude.com/docs/en/about-claude/pricing, checked 2 Oct 2026
context, max output, release, api id, knowledge cutoff, modalities, min cacheable prompt, legacy status: platform.claude.com/docs/en/models/opus-5/overview, checked 2 Oct 2026
status, retirement commitment (not sooner than 2027-07-24): platform.claude.com/docs/en/about-claude/model-deprecations, checked 2 Oct 2026
positioning: anthropic.com/news/claude-opus-5, checked 2 Oct 2026
benchmarks (comparison columns): anthropic.com/claude-opus-5-5, checked 2 Oct 2026
benchmarks (comparison columns): anthropic.com/claude-fable-and-mythos-5-1, checked 2 Oct 2026
output speed: artificialanalysis.ai/models/claude-opus-5, checked 2 Oct 2026

Building on Claude Opus 5?

Get the architecture right before the bill arrives.

We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.

Book a free call All AI models