How much does the Claude Opus 4.8 API cost?

Claude Opus 4.8 lists at $5 per million input tokens and $25 per million output tokens as of August 2026. On a realistic task of 1,500 input and 500 output tokens across 3 calls, that is $0.06000 per task, or $60,000 per million tasks.

Is Claude Opus 4.8 expensive compared to other models?

Claude Opus 4.8 ranks #9 of 10 tracked models on cost per task. It costs 38.1x more than the cheapest tracked model (Mistral Small), while the most expensive (GPT-5.5) costs 1.13x more than it.

When does self-hosting beat the Claude Opus 4.8 API?

Against a $2/hour dedicated GPU ($1,460/month), self-hosting becomes cheaper than Claude Opus 4.8 above 24,333 tasks per month, within that box's physical capacity of 788,400 tasks.

Model pricing · Anthropic · frontier tier · August 2026

Claude Opus 4.8 API pricing, in numbers that matter

List price, real cost per task, rank against the field, and the self-hosting break-even, all computed from our open dataset.

Input, $/1M tokens

Output, $/1M tokens

$25

Cost per task

$0.06000

Cost rank

#9 of 10

The task profile behind every figure: 1,500 input and 500 output tokens per call, 3 calls per task. On that profile Claude Opus 4.8 costs $0.06000 per task and $60,000 per million tasks. Output is billed at 5x the input rate, so despite the profile sending three times more input than output, output makes up 63% of the bill. Capping response length is the highest-return prompt-level optimisation on this model.

Claude Opus 4.8 against the field

#	Model	$/task	vs Claude Opus 4.8	Head-to-head
1	Mistral Small	$0.00158	0.03x	compare
2	DeepSeek-V4-Flash	$0.00396	0.07x	compare
3	Gemini 2.5 Flash	$0.00510	0.08x	compare
4	Grok 4.3	$0.00938	0.16x	compare
5	GPT-5.4 mini	$0.01013	0.17x	compare
6	Claude Haiku 4.5	$0.01200	0.20x	compare
7	Claude Sonnet 5	$0.02400	0.40x	compare
8	Gemini 3.1 Pro	$0.02700	0.45x	compare
9	Claude Opus 4.8	$0.06000	-	-
10	GPT-5.5	$0.06750	1.13x	compare

Self-hosting break-even

Against a reference $2/hour dedicated instance ($1,460/month fixed, 788,400 tasks/month capacity at full utilisation), the break-even against Claude Opus 4.8 is 24,333 tasks per month. That sits inside the box's capacity, so past that volume self-hosting genuinely wins on cost. Work through your own numbers in the $/task calculator or the break-even guide.

Frequently asked

How much does the Claude Opus 4.8 API cost?: Claude Opus 4.8 lists at $5 per million input tokens and $25 per million output tokens as of August 2026. On a realistic task of 1,500 input and 500 output tokens across 3 calls, that is $0.06000 per task, or $60,000 per million tasks.
Is Claude Opus 4.8 expensive compared to other models?: Claude Opus 4.8 ranks #9 of 10 tracked models on cost per task. It costs 38.1x more than the cheapest tracked model (Mistral Small), while the most expensive (GPT-5.5) costs 1.13x more than it.
When does self-hosting beat the Claude Opus 4.8 API?: Against a $2/hour dedicated GPU ($1,460/month), self-hosting becomes cheaper than Claude Opus 4.8 above 24,333 tasks per month, within that box's physical capacity of 788,400 tasks.