Anthropic Drops Claude Opus 5.5: Flagship Frontier Power at 40% Lower Cost

Anthropic has released Claude Opus 5.5, the newest model in its Claude lineup, promising performance close to its top-tier Fable model at a significantly lower price. This is Anthropic's first model release since CEO Dario Amodei publicly called for the AI industry to slow its pace of frontier development.

What's New in Opus 5.5

According to Anthropic, Opus 5.5 is a major step up from its predecessor, Opus 5, with strong gains in coding, computer use, chart and visual recognition, and complex reasoning tasks. Anthropic says the model performs at a level similar to its more capable Fable 5.1 model on most everyday work, while running about 30% faster and roughly 40% cheaper.

The company also says the model communicates more naturally, structuring responses to put the most important information first, and shows improved behavior on internal safety tests, attempting to bypass its own operating boundaries about 85% less often than Opus 5 did in testing.

Pricing

Opus 5.5 costs less to run than the model it replaces:

• Input tokens: $4 per million (down from $5 for Opus 5)
• Output tokens: $20 per million (down from $25 for Opus 5)

Pricing has become an increasingly important competitive factor between AI labs, since more capable models are often also the most expensive to run at scale. Cutting costs while improving performance is a rarer combination.

Availability

Claude Opus 5.5 is available now as a managed API model across the Claude Platform, Amazon Web Services, Google Cloud, and Microsoft Azure. Anthropic has not released the model's weights, so self-hosting is not an option. Zero data retention is available for enterprise customers, consistent with previous Opus releases.

Anthropic also raised usage limits for Pro, Max, Team, and Enterprise subscribers alongside the launch, and gave subscribers a rate limit reset usable through October 22.

Safety Testing

Anthropic says Opus 5.5 was evaluated before release by external testers, including Frontier Design and METR, and scored the best result to date on the company's internal automated behavioral audit, a test covering close to 2,000 scenarios. Because its biology and cybersecurity capabilities are comparable to Anthropic's more restricted Mythos 5.1 model, Opus 5.5 ships with similar safeguards in those areas, including routing most advanced cybersecurity tasks to an older, more restricted model instead.

The Wider Competitive Picture

This release lands in the middle of a broader wave of AI product updates. Microsoft recently redesigned Copilot into an all-in-one "super app" combining chat, coding tools, and autonomous agents, while OpenAI has continued expanding its own agent and coding-focused tools. Reports also suggest OpenAI is preparing new models in its GPT-6 line, though full details have not been officially confirmed.

Coming just three weeks after Anthropic's previous major release, Fable 5.1, some observers have noted the pace of this release sits in some tension with Amodei's own call to slow down frontier AI development, a reminder of how competitive pressure in the AI industry continues even as companies talk publicly about caution.

Frequently Asked Questions

What is Claude Opus 5.5? It's Anthropic's newest AI model, offering performance close to its more capable Fable 5.1 model at a lower price and faster speed than the previous Opus 5.

How much does Opus 5.5 cost? $4 per million input tokens and $20 per million output tokens, down from $5 and $25 for Opus 5.

Is Opus 5.5 available to everyone? Yes, it's available now through the Claude Platform, AWS, Google Cloud, and Microsoft Azure.

Is this Anthropic's most powerful model? No. Anthropic's Fable and Mythos models remain its most capable tier. Opus 5.5 is designed to match Fable-level performance on most everyday tasks at a much lower cost.

Bottom Line

Claude Opus 5.5 shows Anthropic pushing on both performance and price at the same time, a combination that puts pressure on competitors like OpenAI and Google as the race for enterprise and developer AI usage continues to intensify.

Sources: Anthropic, TechCrunch, MacRumors, 9to5Mac, MarkTechPost, Yahoo Finance. Information is accurate as of September 27, 2026 and may be updated as Anthropic releases further models in the 5.5 family.