Anthropic · AI model API pricing

How much does Claude Opus 5 cost?

Anthropic's current Opus tier for complex agentic coding and enterprise work — priced identically to Opus 4.8 while superseding it.

Rate card

What that means per month

Standard on-demand rates from Anthropic's official pricing docs, last checked 2026-07-27. Cached-input reads bill at $0.50 per 1M; cache writes are $6.25 per 1M (5-minute) or $10.00 per 1M (1-hour). Batch API halves the rate to $2.50 in / $12.50 out. Fast mode is a separate premium tier at $10 in / $50 out. Monthly figures exclude caching and batch discounts. Confirm live pricing before committing.

Where Claude Opus 5 sits on price

At the reference month of 10M input and 3M output tokens, Claude Opus 5 costs $125, which makes it the 27th cheapest of the 30 models tracked here. That is 74× the price of Qwen-Flash, the cheapest model at this mix ($1.70). Among the 10 frontier-class models it ranks 7th, behind GLM-5.2 at $27.2.

Cache economics

Cached input on Claude Opus 5 bills at $0.50 per 1M — 10% of its own input rate, against a 10% median across the models here that publish a cached-read price. 2 models discount cached reads more steeply than this. On the reference month, serving 80% of that input from cache takes the input side of the bill from $50 to $14, leaving output untouched at $75 — which is why a high reuse rate changes the ranking above rather than just shaving the total.

Cache reads only. Writes are billed separately by some providers and are excluded here — see the rate note above for Anthropic's terms. Try it at your own hit rate on the AI cost calculator.

Cheaper alternatives

At a reference workload of 10M input / 3M output tokens a month, these cost less than Claude Opus 5 ($125/mo):

• Kimi K3 $75/mo
• Gemini 3.1 Pro $56/mo
• GPT-5.6 Terra $56/mo

Claude Opus 5 head-to-head