llamaperf
Aug 3, 2026
Throughput
28.0 t/s gen
System RAM
192 GB

Summary

User reports decode speeds at various context depths: 28 t/s at the start, 23.5 t/s at 45k, and 18 t/s at 192k. The run was maintained with 8k token output. Prefill performance is mentioned but no numbers are given.