Anthropic Releases Claude Opus 4.8, Reclaiming the Coding Crown

Anthropic has released Claude Opus 4.8, its most capable model, delivering leading results on agentic coding and long-horizon reasoning tasks with a one-million-token context window. The launch is accompanied by Fable 5, which reaches 80.3% on the demanding SWE-bench Pro benchmark, and an updated Sonnet 5 for balanced everyday use. Opus 4.8 is priced at $5 input and $25 output per million tokens.