Anthropic released Claude Opus 5.5 on Tuesday, the first model in what the company is calling its Claude 5.5 family. The model is priced at $4 and $20 per million input and output tokens—40% cheaper than Opus 5 at default settings—and Anthropic says it matches Claude Fable 5.1, its priciest flagship, on most tasks.

On Anthropic's own numbers, Opus 5.5 leads both Fable 5.1 and its predecessor, Opus 5, on coding and knowledge-work tests, though GPT-6 Astra still beats it on AutomationBench and Terminal-Bench-Science 0.1. The new model launches with the same cybersecurity and biology safeguards as Fable 5.1.

The release undercuts both the model it replaces and the flagship it chases: Opus 5.5 costs 20% less to run than Opus 5 and is 60% cheaper than Fable 5.1, the company's state-of-the-art offering.

The model is Anthropic's most advanced by both version (5.5 versus Fable's 5.1) and family tier—Opus the largest model publicly available, followed by Sonnet, with Haiku the smallest. Anthropic acknowledges that the real-world gap between Fable and Opus is narrower than benchmark scores suggest, but Opus's public availability on paid plans and its lower per-token cost make it the obvious choice for most Claude users.

The two product lines have traded the lead this year. Opus 5 launched in July, underpricing and outscoring Fable 5 on most benchmarks. Fable 5.1 arrived in early September and reversed that, beating Opus 5 on every benchmark Anthropic published. Opus 5.5 closes the gap again—this time on price rather than raw scores.

Lovable, a developer platform that ran Opus 5 through its own coding tests in July, said in a statement shared by Anthropic that the earlier model was "steadier, with far less variance run to run" than what came before it—the same efficiency pitch Anthropic is making again now.

Opus 5.5 is also the first model Anthropic has shipped since CEO Dario Amodei published an essay calling on the industry to slow down, arguing that AI capability gains are outrunning the safety testing meant to keep them in check. "We must slow the pace at which we improve the capabilities of AI models," Amodei wrote earlier this month. Anthropic has now shipped three flagship-class models since the summer—Opus 5 in July, Fable 5.1 in early September, and Opus 5.5 on Tuesday—the pace the essay argues should slow.

Anthropic measures much of this progress through agentic coding—work carried out by an AI agent, a model that takes multi-step actions on its own rather than simply answering a single prompt.

On Terminal-Bench 4.0, which tests whether an agent can complete complex professional tasks inside a command-line interface and scores results as a percentage of tasks completed, Opus 5.5 hit 66.4%, ahead of Fable 5.1's 55.8% and GPT-6 Astra's 57.9%. FrontierCode, which checks whether an agent's code changes would actually be merged in a real engineering pipeline using the same pass-rate scoring, put Opus 5.5 at 54.4% against Fable 5.1's 50.3%.

On GDPval-AA v2.1, which grades real-world professional work across 44 occupations using Elo—the chess-style ranking system that measures relative skill rather than a flat percentage—Opus 5.5 scored 1846, against Fable 5.1's 1735 and Opus 5's 1708.

Opus 5.5 does not lead everywhere. On AutomationBench, a Zapier-built benchmark that scores whether an agent can carry a full business workflow from start to finish without human help, GPT-6 Astra's 41.4% edges out Opus 5.5's 40.0%. On Terminal-Bench-Science 0.1, which tests agentic scientific research conducted inside a command-line environment using the same pass-rate method, Astra's 64.6% beats Opus 5.5's 58.7% by a wider margin.

The bigger selling point is cost. Input tokens—the chunks of text a model reads and writes, priced per million—now run $4 for Opus 5.5 versus $5 for Opus 5, and output tokens drop to $20 from $25. Cache reads, which reuse context a model has already processed instead of reprocessing it from scratch and drive most of the cost of long agentic coding sessions, fall 60% to $0.20 per million tokens. The structure makes the discount compound where it counts: extended agentic coding sessions are billed mostly on cache reads, so the steepest cut lands on the largest line item.

Because Opus 5.5 matches Anthropic's Mythos-class models—the company's most restricted tier, reserved for vetted cybersecurity and biology researchers—on those two capabilities, it launches with the same safeguards as Fable 5.1. Most cybersecurity tasks are automatically rerouted to the older Opus 4.8, a practice many developers and users have previously complained about. It remains to be seen whether Opus 5.5's arrival changes that rerouting behavior.

Opus 5.5 is live now across Anthropic's platforms, including Amazon Web Services, Google Cloud, and Microsoft Azure. Sonnet 5.5 and Haiku 5.5 will follow in the coming weeks, Anthropic says, carrying the same price cuts down the rest of the lineup.