Anthropic Releases Claude Opus 5 at Opus 4.8 Pricing
Anthropic released Claude Opus 5 on July 24 at $5 per million input tokens and $25 per million output tokens, the same base pricing as Opus 4.8 and half the price of Claude Fable 5. The company says Opus 5 approaches Fable 5 on many coding and knowledge-work tasks, while Axios reports it is becoming the default model for Claude Max subscribers.
Anthropic released Claude Opus 5 on July 24 across its platforms and API. The model costs $5 per million input tokens and $25 per million output tokens, matching Opus 4.8's base pricing and coming in at half the price of Claude Fable 5.
Anthropic positions Opus 5 as an everyday model for coding, knowledge work and scientific research. It is the new default for Claude Max and the strongest model available on Claude Pro. Axios separately reported the launch and the model's placement across Anthropic's paid plans.
What Anthropic's evaluations show
The company says Opus 5 more than doubled Opus 4.8's performance on Frontier-Bench v0.1 at a lower cost per task. On CursorBench 3.2, Anthropic reports that the model at maximum effort came within 0.5% of Fable 5's peak score while costing half as much per task. It also reports gains on automation, computer-use and life-sciences evaluations.
Those figures are vendor-reported results, not independent validation. Anthropic's Frontier-Bench comparison used the mini-SWE-agent harness, a Google Kubernetes Engine backend and five attempts per task. Teams comparing models should therefore reproduce results with their own prompts, tools, latency targets and failure costs.
Opus 5 also adds effort controls that let users trade more inference work for higher task performance. Anthropic says a separate Fast mode runs about 2.5 times faster than the default mode at twice the base price.
Capability and safety boundaries
Anthropic says Opus 5 remains behind Claude Mythos 5 on biology research and offensive cybersecurity. The company also says the model's cyber safeguards block binary vulnerability scanning, penetration testing and exploit generation, while allowing more source-code vulnerability work than the stricter Fable 5 controls.
For production teams, the practical comparison is not a single benchmark rank. Opus 5's launch creates another point on the quality-cost curve, so evaluation should combine task success, token use, latency, tool reliability and safety behavior before a workload is moved from Opus 4.8 or Fable 5.
Key Points
- 1Claude Opus 5 launched July 24 at $5 per million input tokens and $25 per million output tokens, the same base pricing as Opus 4.8.
- 2Anthropic reports near-Fable performance on several coding and knowledge-work tests, but the benchmark results are vendor-reported.
- 3Effort controls and a higher-priced Fast mode give teams additional ways to trade quality, latency and inference cost.
Scoring Rationale
Claude Opus 5 is a major model release with materially lower pricing than Anthropic's top tier and direct relevance to production coding agents, evaluation strategy and inference economics.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
