Anthropic has officially released the fast mode for its Claude Opus 5.5 model, removing the previous preview access requirements. This new mode offers up to 2.5 times higher output tokens per second but comes at double the cost of the standard Opus 5.5 model. The speed enhancement primarily benefits output token generation, not the time to the first token, and switching between fast and standard modes can invalidate the prompt cache, potentially increasing costs for certain workflows. AI
IMPACT Accelerates inference for specific workloads, but at a higher cost and with potential cache invalidation issues.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →