Sam's News — anthropic — 2026-09-29¶

Anthropic¶

7 Anthropic releases Claude Opus 5.5, claiming major performance gains and 40% cost reduction¶

Anthropic introduced Claude Opus 5.5, a new flagship model that matches Claude Fable 5.1 performance while costing 40% less to run. The model underwent external safety testing and scored highest on Anthropic's alignment audit. A companion model, Claude Sonnet 5.5, was released days later with 30% speed improvements and lower costs.

Claude Opus 5.5 costs 40% less than Opus 5 with input tokens at $4 per million and output at $20 per million
Opus 5.5 generates output 30% faster than Opus 5
Cache read costs dropped 60% to $0.20 per million tokens
Model was tested by external evaluators including Frontier Design and METR before release
Opus 5.5 achieved the strongest scores to date on Anthropic's automated behavioral audit for alignment testing
Claude Sonnet 5.5 followed within a week, running 30% faster and costing up to 30% less than Sonnet 5

Sources: Anthropic AI Web Searched, Mashable AI Web Searched