Anthropic introduces Claude Opus 5.5

Anthropic has introduced Claude Opus 5.5 — the first model in the Claude 5.5 family, which outperforms Opus 5 in agentic programming, computer use, and knowledge work, operating 30%+ faster and approximately 40% cheaper on typical tasks.

Compiled by Sergey KostenchukPublished 2026-09-22Updated 2026-09-22

2026-09-22 Research Anthropic

🤖 Anthropic introduces Claude Opus 5.5

The first model in the 5.5 family outperforms Opus 5 in agentic programming, computer use, and knowledge work — 30%+ faster and approximately 40% cheaper on typical tasks. API prices: $4/$20 per 1M tokens compared to $5/$25, cache reading — $0.20. Terminal-Bench 4.0 — 66.4%, FrontierCode v1.1 — 54.4%, Humanity's Last Exam — 67.7%.

🌍 The flagship level has become cheaper for the first time: input $4 instead of $5, output $20 instead of $25, cache 60% cheaper — this directly reduces the cost of agentic pipelines and long-running tasks. In Claude Code and on the Platform, there is a Fast mode up to 2.5x faster at $8/$40 per 1M.

👤 For work in Claude Code and via API, the model responds more concisely, is more resistant to prompt injection, and limits for Pro, Max, Team, and Enterprise have been increased. Sonnet 5.5 and Haiku 5.5 — in the coming weeks.

Source 1: https://www.anthropic.com/claude-opus-5-5

Sources