Claude Opus 5
We recommend using Claude Opus 5.5. It replaces Opus 5 with a higher CursorBench score at 20% lower input and output rates.
Claude Opus 5 is Anthropic's previous Opus model and replaces Opus 4.8. It is a step-change over Opus 4.8 on agentic coding, professional knowledge work, and long-horizon reasoning, and it lands on par with Fable 5 on CursorBench at Opus pricing. We recommend the high thinking variant for the best results.
Strengths
- Stronger than Opus 4.8 on thoroughness, carefulness, and intent understanding. It holds the goal across long sessions and finishes multi-step work with less hand-holding.
- Competitive with Fable 5 on hard coding work, often finishing tasks faster end-to-end while charging standard Opus rates.
- Strong at planning and tool use. It maps work before executing, chains tool results into follow-up actions, and adapts when tool output surprises it.
- Zero Data Retention compatible. Unlike Fable 5 and Fable 5.1, Opus 5 does not require Anthropic data-retention opt-in.
Limitations
- Costs more than Opus 5.5 per token ($5/$25 vs $4/$20) while scoring lower on CursorBench.
- Same per-token price as Opus 4.8, so it still uses prepaid credit (or the Other Models pool on legacy Cursor plans) faster than Sonnet or Composer.
- Can over-elaborate in long sessions where brevity matters more than depth.
Tools
Opus 5 has access to all agent tools when used with Cursor including:
Find files by name, read directory structures, and grep for patterns.
Generate search queries and fetch results from the web.
Read file contents, including images for vision-capable models.
Suggest edits and apply them automatically.
Execute terminal commands and monitor output.
Control a browser to take screenshots, test applications, and verify visual changes. See the Browser documentation.
Generate images from text descriptions or reference images.
Ask clarifying questions while continuing to work in the background.
Retrieve rules based on type and description.
Learn more about how tools work and tool calling fundamentals.
Pricing
Opus 5 is a third-party model, so it isn't part of the usage included with Cursor plans. On The X Subscription it's billed at the rates below from your prepaid credit. On legacy Cursor plans it draws from the Other Models pool. All prices are per million tokens.
| Name | ||||
|---|---|---|---|---|
| $5 | $6.25 | $0.5 | $25 | |
| $10 | $12.5 | $1 | $50 |
A Fast mode tier (claude-opus-5-fast) is available at launch for higher-priority output. On legacy request-based plans, it requires Max Mode. It bills at $10/M input and $50/M output tokens. Use it selectively for time-sensitive or critical work.
All Opus 5 prompts bill at the base per-token rates in the table above, including when context goes above 300k. There is no separate long-context multiplier for Opus 5. Context windows up to 1M tokens use the same rates.
Opus 5 supports a thinking variant for deeper reasoning. We recommend using the high thinking variant for the strongest results.