AnthropicmodelClaude Opus 5
Anthropic's new top model, available today on all paid Claude plans and the API at the same price as Opus 4.8.
Details
- Costs $5 per million input tokens and $25 per million output tokens, the same as Opus 4.8
- Replaces Opus 4.8 as the default model on Claude Max and is the strongest option on Claude Pro
- Scores within 0.5% of Fable 5 on the Frontier-Bench and CursorBench software engineering evaluations at half the cost per task
- Beats Fable 5 on OSWorld 2.0, a computer-use test, at roughly a third of the cost
- Runs in Fast mode at 2.5x the default speed for double the price
- Available on Claude.ai, Claude Code, Claude Cowork, and the Claude API
- Stronger than Opus 4.8 on cybersecurity tasks, with safeguards meant to allow vulnerability fixing while blocking exploit development
Best for
Long agentic coding runs and multi-step analysis where you want the model to check its own work instead of handing back a first pass.
When to use it
You are on Claude Pro or Max and want the strongest model available, or you are paying Fable 5 rates on the API for work Opus 5 handles at half the cost per task.
Who benefits
General Claude users + teams running Claude Code or Cowork, plus API users watching per-task cost.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →