Breaking

Anthropic releases Claude Opus 5.5 with lower pricing, higher limits and Fable-level performance

September 23, 2026 at 00:46 EDT

Foundation Models
AI Agents
Software Dev & Coding

Anthropic released Claude Opus 5.5 on September 22, positioning the first model in its Claude 5.5 family as a cheaper, faster upgrade for agentic coding and knowledge work. The company says the model performs at roughly the level of Claude Fable 5.1 for most work while costing 40% less to run than Opus 5, and it is pairing the launch with higher five-hour usage limits and a bankable rate-limit reset for paid Claude users.

Sep. 22, 2026 · Anthropic

Opus 5.5 resets the Claude cost curve

Anthropic says its new model reaches Fable 5.1-level work for most tasks while cutting typical Opus 5 running costs by 40% and raising paid usage limits.

40%

lower typical cost than Opus 5

66.4%

Terminal-Bench 4.0 result

token context window

30%+

faster output than Opus 5

Typical running cost index

Anthropic frames Opus 5.5 as the same workload for 60 units of cost where Opus 5 was 100.

100

Opus 5

Opus 5.5

Why builders like it

Lower token prices, clearer summaries, higher five-hour limits and a saveable reset make long Claude Code sessions easier to justify.

What still needs care

Per-task savings vary, always-on thinking can change token use, and high-risk cyber or biology work may be routed through safeguards.

Default path for teams

Start on Opus 5.5→Measure real sessions→Escalate to Fable 5.1 for hardest runs

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.