Anthropic has launched Claude Opus 5, a hybrid reasoning model that more than doubles the performance of its predecessor at the same price.
Anthropic has introduced Claude Opus 5, an update to its hybrid reasoning model that brings improved performance on coding and knowledge tasks, more capable AI agents, and sharper output for professional work.
What’s New in Claude Opus 5
Introduced on July 24, Claude Opus 5 delivers substantially improved performance at the same cost as its predecessor, Opus 4.8, according to Anthropic. The company described the new model as thoughtful and proactive, positioning it close to the frontier intelligence of Claude Fable 5 while costing half as much.
Claude Opus 5 is now the default model on Claude Max and the strongest model available on Claude Pro, reflecting Anthropic’s push to make its most capable reasoning available across its consumer tiers.
- Same pricing as Opus 4.8 despite major performance gains
- Default model on Claude Max
- Strongest model on Claude Pro
- Positioned close to Fable 5’s frontier performance at lower cost
Benchmark Performance
Anthropic pointed to specific benchmark results to illustrate the upgrade. On Frontier-Bench v0.1, Claude Opus 5 surpasses all other models and more than doubles Opus 4.8’s performance while costing less per task.
On CursorBench 3.2, at max effort settings, Opus 5 performs within 0.5% of Fable 5’s peak score, but at half the cost per task. Anthropic also said Opus 5 achieves greater performance for a given cost than all other models across high, xhigh, and max effort settings.
| Benchmark | Result |
|---|---|
| Frontier-Bench v0.1 | More than double Opus 4.8’s performance at lower cost per task |
| CursorBench 3.2 (max effort) | Within 0.5% of Fable 5’s peak score, at half the cost |
Pricing and API Access
Claude Opus 5 is priced at $5 per million input tokens and $25 per million output tokens, matching the pricing of Opus 4.8. Developers can access Claude Opus 5 through the Claude API.
By keeping pricing unchanged while improving performance, Anthropic is positioning Opus 5 as a straightforward upgrade path for existing developers and businesses already building on Opus 4.8, without requiring them to absorb higher costs.
Additional Platform Updates
Alongside the release of Opus 5, Anthropic also rolled out two features in beta:
- Mid-conversation tool changes on the Claude Platform
- Automatic fallbacks on the API
These beta features are aimed at improving flexibility for developers building applications on Anthropic’s platform, allowing tools to be swapped during an active conversation and providing automatic recovery options if a request encounters issues.
What This Means for Users and Developers
For everyday users, the shift to Opus 5 as the default model on Claude Max means access to stronger reasoning and coding capabilities without any change in subscription cost. For developers, the unchanged API pricing paired with significant performance gains offers a way to improve application quality without increasing operating expenses.
- Claude Max users get Opus 5 as their new default model automatically
- Claude Pro users gain access to Anthropic’s strongest available model
- Developers can adopt Opus 5 via the API at the same price as before
Frequently Asked Questions
Claude Opus 5 is Anthropic's updated hybrid reasoning AI model, offering improved coding and knowledge task performance at the same cost as its predecessor, Opus 4.8.
Claude Opus 5 is priced at $5 per million input tokens and $25 per million output tokens, the same as Opus 4.8.
Yes, Claude Opus 5 is the new default model on Claude Max and the strongest model available on Claude Pro.
