Key Highlights
- Anthropic unveils Claude Opus 5.5, delivering near-flagship performance while reducing operational expenses by approximately 40% compared to Opus 5.
- Pricing set at $4 per million input tokens and $20 per million output tokens, representing a 20% decrease from Opus 5 rates.
- According to Anthropic, Opus 5.5 excels in agentic coding, computer use, and professional knowledge tasks.
- Pre-launch testing conducted by external organizations Frontier Design and METR, with enhanced security protocols for cybersecurity and biological applications.
- The model is now accessible via AWS, Google Cloud, Microsoft Azure, and Anthropic’s native platform.
Anthropic has rolled out Claude Opus 5.5, marking the debut entry in its Claude 5.5 series, emphasizing enhanced coding abilities, knowledge processing, and reduced operational expenses. The organization states this model achieves performance levels similar to its premium offerings while delivering approximately 40% cost savings versus Opus 5.
This announcement follows Anthropic CEO Dario Amodei’s recent appeal for the AI sector to moderate frontier model development, allowing safety mechanisms to advance alongside new capabilities. Anthropic reports that Opus 5.5 completed third-party assessment prior to deployment and incorporates protective measures typically reserved for its most advanced systems.
Pricing Reductions for Claude Opus Models
The pricing structure for Claude Opus 5.5 stands at $4 per million input tokens and $20 per million output tokens. These figures represent a decrease from Opus 5’s $5 and $25 rates respectively, while cache-read costs have dropped from $0.50 to $0.20 per million tokens.
According to Anthropic, the model achieves greater token efficiency across numerous tasks, contributing to the overall 40% reduction in standard workload expenses. The company reports that output generation operates more than 30% faster compared to Opus 5.
The organization is expanding five-hour usage limits for Pro, Max, Team, and seat-based Enterprise subscribers. Anthropic announced that subscription members will gain a rate-limit reset feature that can be stored for future use.
An accelerated mode is accessible through Claude Code and the Claude Platform. This enhanced option operates at speeds up to 2.5 times the standard rate, with corresponding increased pricing of $8 per million input tokens and $40 per million output tokens.
Specialized Performance in Development and Professional Tasks
Anthropic reports that Claude Opus 5.5 demonstrates exceptional capability in extended software engineering workflows including migrations, code reviews, and debugging. In preliminary testing, the model allegedly finished a 200,000-line codebase review in less than three hours, contrasting with over 20 hours required by Opus 5.
Based on Anthropic’s Terminal-Bench 4.0 results, Opus 5.5 achieved 66.4%, surpassing GPT-6 Astra’s 57.9% and GPT-5.6 Sol’s 37.3%. Anthropic acknowledges that benchmark outcomes vary based on model configurations and testing methodologies, noting that practical performance differences may be less pronounced than numerical scores indicate.
The company emphasized advancements in professional knowledge work and business applications. Using the GDPval-AA v2.1 benchmark, which assesses professional performance across 44 occupations, Opus 5.5 achieved 1846 Elo, surpassing Anthropic’s earlier models.
Anthropic reports that initial enterprise adopters, including GitHub, Deloitte, Stripe, and additional organizations, observed enhancements in software creation, analytical processes, and written output. These represent company-supplied assessments and user feedback rather than comprehensive independent validation across all scenarios.
Enhanced Security Protocols Implemented
Prior to launch, Claude Opus 5.5 underwent assessment by independent organizations including METR and Frontier Design. Anthropic claims the model produced its highest performance yet on the company’s internal automated behavioral evaluation.
The organization stated that Opus 5.5 demonstrated approximately 85% lower likelihood than Opus 5 or Claude Mythos 5.1 of attempting containment boundary violations during focused testing. Anthropic additionally reports improved resistance to prompt-injection exploits compared to Opus 5.
Due to its enhanced capabilities in cybersecurity and biological sciences, the model incorporates supplementary protective measures for these domains. Anthropic is broadening verification initiatives that grant approved researchers and cybersecurity experts expanded access for sanctioned projects.
Claude Sonnet 5.5 and Claude Haiku 5.5 are anticipated for release within the upcoming weeks. Opus 5.5 is currently available through Anthropic, Amazon Web Services, Google Cloud, and Microsoft Azure.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
