Claude Opus 4.8

Anthropic's most capable model for deep reasoning, coding, and long context.

Provider	Anthropic
Context window	1,000,000 tokens (1.0M)
Input price	$5.00 / 1M tokens
Output price	$25.00 / 1M tokens
Inputs	Text, Images
Reasoning mode	Yes
Vision (image input)	Yes
Tool / function calling	Yes
Native real-time web	No
Open-weight	No

Pricing & context web-verified June 2026 — confirm on Anthropic's pricing page. Supports a 1M-token context window.

What Claude Opus 4.8 is best for

Complex reasoning and analysis
Large-codebase and agentic coding
Long-document synthesis
High-stakes writing

How to prompt Claude Opus 4.8

Use XML-style tags (<context>, <task>, <format>) — Claude follows tagged structure very literally.
Turn on extended thinking for hard problems, and ask it to plan before it answers.
Give explicit constraints ("do not…", "limit to…") — it respects them closely.
It handles a full 1M-token context window — paste whole codebases or document sets rather than chunking.

Work with Claude Opus 4.8

Build an optimized prompt with the Anthropic prompt generator.
Estimate cost with the token counter & cost calculator, or compare every model on the AI model comparison page.
Related reading: Best Claude Opus 4 7 Prompts 2026 · Claude Opus 4 7 Prompting Guide.
Official: claude.ai

Frequently asked questions

What is Claude Opus 4.8's context window?

Claude Opus 4.8 has a 1,000,000-token context window (about 1.0M tokens) — the maximum amount of text it can consider at once.

How much does Claude Opus 4.8 cost?

Via the Anthropic API, Claude Opus 4.8 costs $5.00 per 1M input tokens and $25.00 per 1M output tokens (web-verified June 2026). Output tokens cost more, so the length of the model's reply usually drives the bill.

Is Claude Opus 4.8 multimodal?

Yes — Claude Opus 4.8 accepts Text, Images as input, not just text.

Is Claude Opus 4.8 open-weight?

No — Claude Opus 4.8 is available only through the Anthropic API, not as downloadable weights.

Claude Opus 4.8 specifications