Anthropic Launches Claude Opus 5.5 With Lower Costs and Stronger Coding Performance

TLDR

• Anthropic has launched Claude Opus 5.5, saying it delivers performance close to its top-tier models while costing about 40% less to run than Opus 5.
• The model costs $4 per million input tokens and $20 per million output tokens, both 20% below Opus 5 pricing.
• Anthropic says Opus 5.5 leads in areas including agentic coding, computer use and knowledge work.
• The model was externally tested by Frontier Design and METR before release and includes stronger safeguards for cybersecurity and biology.
• Claude Opus 5.5 is available through AWS, Google Cloud, Microsoft Azure and Anthropic’s own platform.

Anthropic has launched Claude Opus 5.5, the first model in its new Claude 5.5 family, with a focus on stronger coding, knowledge work and lower operating costs. The company says the model performs at a level comparable with its highest-end systems while costing about 40% less to run than Opus 5.

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.

It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5. pic.twitter.com/Q9C2VKQ79f

— Claude (@claudeai) September 22, 2026

The release comes shortly after Anthropic CEO Dario Amodei called for the AI industry to pace frontier development so safety systems can keep up with new capabilities. Anthropic says Opus 5.5 underwent external evaluation before launch and includes safeguards previously reserved for its most capable models.

Anthropic Cuts Claude Opus Pricing

Claude Opus 5.5 is priced at $4 per million input tokens and $20 per million output tokens. That compares with $5 and $25 respectively for Opus 5, while cache-read pricing has fallen from $0.50 to $0.20 per million tokens.

Anthropic says the model also uses fewer tokens to complete many tasks, which contributes to the overall 40% reduction in typical workload costs. Output generation is more than 30% faster than Opus 5, according to the company.

The company is also increasing five-hour usage limits for Pro, Max, Team and seat-based Enterprise customers. Anthropic said subscription users will receive a rate-limit reset that can be saved and used later.

A faster mode is also available through Claude Code and the Claude Platform. That option can run at up to 2.5 times normal speed, with higher pricing of $8 per million input tokens and $40 per million output tokens.

Opus 5.5 Targets Coding and Knowledge Work

Anthropic says Claude Opus 5.5 performs particularly well on long software engineering tasks such as migrations, code audits and bug fixing. In one early test, the model reportedly completed a 200,000-line codebase audit in under three hours compared with more than 20 hours for Opus 5.

On Anthropic’s reported Terminal-Bench 4.0 results, Opus 5.5 scored 66.4%, compared with 57.9% for GPT-6 Astra and 37.3% for GPT-5.6 Sol. Anthropic notes that benchmark results depend on model settings and evaluation methods, and says real-world differences can be narrower than raw scores suggest.

The company also highlighted improvements in knowledge work and business tasks. On the GDPval-AA v2.1 benchmark, which evaluates professional work across 44 occupations, Opus 5.5 recorded a score of 1846 Elo, ahead of Anthropic’s previous systems.

Anthropic says early enterprise testers including GitHub, Deloitte, Stripe and other companies reported improvements in software development, analysis and written communication. These are company-provided evaluations and customer reports rather than independent measurements of every real-world use case.

Anthropic Adds Stronger Safety Controls

Claude Opus 5.5 was evaluated before release by external groups including METR and Frontier Design. Anthropic says the model achieved its strongest results yet on its internal automated behavioral audit.

The company said Opus 5.5 was around 85% less likely than Opus 5 or Claude Mythos 5.1 to attempt to bypass containment boundaries in a dedicated evaluation. Anthropic also says the model is more resistant to prompt-injection attacks than Opus 5.

Because of its capabilities in cybersecurity and biology, the model includes additional safeguards in those areas. Anthropic is expanding verification programs that allow vetted researchers and cybersecurity professionals broader access for approved work.

Claude Sonnet 5.5 and Claude Haiku 5.5 are expected to follow in the coming weeks. Opus 5.5 is available now through Anthropic, Amazon Web Services, Google Cloud and Microsoft Azure.