#Throughput — Kunal Ganglani

gpu server rack datacenter nvidia — illustration for article on Mercury 2.5 770 tok/s Benchmark: The

Mercury 2.5 770 tok/s Benchmark: The Production Playbook

Mercury 2.5 is clocking ~780 tok/s on Artificial Analysis. At that speed, decode stops being your bottleneck. Here’s what actually changes in production: batching, streaming UX, context packing, cost controls, and evals that don’t lie.

Keep exploring

Close to this topic, not on this page.

RAG Evaluation Metrics for Retrieval Quality: My Production Playbook If your RAG app got worse after an embeddings or chunking change, grading answers won’t tell you why. 11 min read
Prompt Compression Token Savings Benchmark: RTK vs Reality Token savings are easy to market and hard to bank. 8 min read
Agent Evaluation Roadmap for Small Teams: 30 Min a Week A pragmatic map of offline vs online vs HITL evals, what to measure beyond “response quality,” and the smallest program that actually… 8 min read