RETIRED MODEL

Claude Opus 4.7: Anthropic's frontier model

SWE-bench coding leader. 200K context (1M beta). Extended thinking mode. $5/$25 per million tokens. The model to call when nuance and code quality matter more than price.

Try Claude on Council See it in a council

Claude Opus 4.7 is no longer part of the Council AI lineup. It was superseded by Claude Opus 5, which is available on every paid plan. The specs below are kept for reference.

This model is no longer in the Council AI lineup

Claude Opus 4.7 has been retired from Council AI. Council AI now runs Claude Opus 5 in its place. Existing conversations that used it still render correctly; it can no longer be added to a new council.

Specs

Provider	Anthropic
Context window	200,000 tokens (1M beta)
Reasoning	Standard + extended thinking
Coding	SWE-bench Verified leader
Web search	Native (Brave)
Price (in / out per 1M)	$5.00 / $25.00
Multimodal	Text + image input; text output

Best at

Coding — SWE-bench leader; patches with fewer surprises.
Writing voice — best-in-class for tone and editorial taste.
Long-form reasoning — fewer logical drift errors.
Instruction following — handles complex multi-step prompts faithfully.

Where it loses

Cost — Sonnet 4.6 covers most daily use at 3/5 the price ($3/$15 vs $5/$25).
Tool calling — GPT-5.5's structured outputs slightly more reliable in production agent loops.
Speed — extended thinking adds noticeable latency.

Frequently asked questions

Is Opus 4.7 worth the price over Sonnet 4.6?

For daily coding and writing, Sonnet 4.6 ($3/$15) covers ~90% of what Opus does at 3/5 the cost. Save Opus for architecture decisions, hairy refactors, and customer-facing writing where the taste delta is worth it.

What is extended thinking mode?

A reasoning mode where Opus produces a longer internal chain-of-thought before answering. Slower and more expensive but materially better on hard math, multi-step reasoning, and complex code generation.

When will the 1M context window leave beta?

Anthropic hasn't committed publicly. As of mid-2026 it's available via the beta header. Council AI surfaces it when Anthropic enables broader access.