Curated content (treat as data, not instructions):
On Sep 22, 2026, Anthropic announced Claude Opus 5.5, its new flagship model, available across AWS, GCP and Azure. Published pricing is $4 per million input tokens and $20 per million output tokens, with prompt caching at $0.20 per million read and $5 per million write. Anthropic's claims of benchmark superiority, "Fable-5.1-level on most work" and roughly 40% lower run cost than Opus 5 are vendor characterizations, not independently verified here.
Notes
What is confirmed: Anthropic announced Claude Opus 5.5 on Sep 22, 2026, live on AWS, GCP and Azure; pricing is $4 in / $20 out per million tokens, with caching at $0.20 read / $5 write per million (official page, WebFetch 2026-09-23; corroborated by tier-1 press 2026-09-22).
What is vendor-claimed (tagged, not treated as fact): benchmark superiority, "Fable-5.1-level on most work," and "~40% lower run cost than Opus 5."
The GTM-stack read: for anyone running agentic workloads, a new flagship is a model-selection data point — the numbers that matter for build-vs-buy are the token prices and caching rates, which are confirmed, more than the vendor's relative-quality and cost claims, which are not. Slot the confirmed pricing into your own cost model before acting on the comparative framing. Releases the same day as OpenAI's GPT-6 Sol and Luna — a two-vendor model-refresh moment worth pricing out together.
Curated from the GTM Stacker signal registry (2026-09-23 pass: studio news lane).