Mercury 2.5 LLM hits 770 tokens per second
Summary
Mercury 2.5 is a proprietary LLM from Inception released in September 2026. It delivers high speed (770.4 tokens per second) and a large context window (~260k tokens) with competitive input/output pricing, but its intelligence benchmark is below average (12/175). The article includes a benchmark breakdown, context window details, and an FAQ that helps compare it to peers for speed-focused use cases.