Large language model
Verdict
Anthropic's July 2026 workhorse flagship. Intelligence 51 on Index v4.3 (behind Fable 5.1 / Astra). Same $5/$25 as Opus 4.8, with a Fast mode at 2×. Still #2 on Arena Agent. The cheaper Claude when you do not need Fable 5.1.
“Still a top coding agent. No longer the Index leader.”
Use when Fable 5.1 is too expensive; route routine tokens to Sonnet 5 or GLM-5.3 Flash.
Training data policy: API data not used for training by default (opt-out / ZDR available)
DPDP Act (India): No India-resident inference by default. Use DPA + region controls.
Data residency: US, EU
API available globally