Claude Opus 4 API Pricing

Anthropic · context window 200K · prices updated 2026-09-15

Input per 1M tokens

$15.00

Output per 1M tokens

$75.00

Cached input per 1M

$1.50

90% cheaper than fresh input

Where Claude Opus 4 sits on price

At $75.00 per million output tokens, Claude Opus 4 is a frontier tier model — 36.1× the median output price of $2.08, which puts it in the most expensive 10% of everything we track. It is the most expensive of the 27 Anthropic models we track. Output costs 5.0× input, the usual spread — trim both, starting with the answer length.

Output tokens per $1

13,333

One full context fill

$3.00

Cheaper than

4% of tracked models

Cached input is 90% cheaper. On an input-heavy workload you need roughly a 28% cache-hit rate to take 25% off the input line — reachable for chatbots and agents that resend the same system prompt and history.

What real workloads cost on Claude Opus 4

These three workloads are the ones teams actually run on a frontier-priced model — a $75.00/1M model is not bought for the same job as one ten times the price.

Workload	Per request	Per month
Small app — 1K requests/day (500 in / 150 out) The baseline most side projects actually run at.	$0.0187	$563
Coding agent — 300 sessions/day (60K in / 12K out) Multi-step loops re-read the same files; caching matters more than raw price.	$1.80	$16200
Support chatbot — 500 conversations/day (5K in / 1.4K out) Conversation history re-sent each turn — the classic prompt-caching win.	$0.18	$2700

Model your exact traffic in the API cost calculator — it preloads Claude Opus 4 with caching and batch options.

Claude Opus 4 price history

Claude Opus 4's price has not moved since we began tracking it on Aug 2, 2026 — 44 days of stability in a market where 78 of the 334 models we track have repriced over the same period. Anthropic's lineup has held steady across the board.

Change-points from our daily price snapshots (tracking since Aug 2, 2026; intraday moves between snapshots are not captured).

Claude Opus 4 vs o3 Pro

The closest-priced alternative from another vendor is o3 Pro (OpenAI) — 7% more expensive on output, with $5.00 more per million input tokens. When two models land this close on price, the decision is quality on your own workload, not the price sheet: run 50 real requests through both and compare.

See o3 Pro pricing →

Cheaper alternatives

GPT-4 OpenAI $30.00 in · $60.00 out 20% less per output token Claude Fable 5.1 Anthropic $10.00 in · $50.00 out 33% less per output token Kimi K3 (batch) Moonshot (Kimi) $3.00 in · $15.00 out 80% less per output token

More Anthropic models

Claude Opus 4.1 Claude Fable 5.1 Claude Fable 5 Claude Opus 4.1 (batch)

Frequently asked questions

› How much does the Claude Opus 4 API cost?

Claude Opus 4 costs $15.00 per million input tokens and $75.00 per million output tokens, with cached input at $1.50 per million (90% cheaper). That works out to roughly 13,333 output tokens per dollar.

› What does the small app workload cost on Claude Opus 4?

Small app — 1K requests/day (500 in / 150 out) costs about $0.0187 per request and $563 per month on Claude Opus 4. The baseline most side projects actually run at.

› What does it cost to fill Claude Opus 4's 200K context window?

Sending 200K of input in a single request costs $3.00 at $15.00 per million tokens — before any output. With prompt caching that same fill drops to about $0.3 on repeat requests. This is why large context windows are cheap to advertise and expensive to actually use.

› Is Claude Opus 4 worth the price?

Claude Opus 4 is priced in the top bracket, at $75.00 per million output tokens. It only pays off where an error is expensive — production code, legal and medical drafting, agents that take real actions. For anything routine, GPT-4 at $60.00 does the same job for a fraction of the bill.