strata serve: prompt 4022 tokens = 0 reused + 4022 read in 12890 ms (312.0 tok/s), 256 generated in 19572 ms (13.1 tok/s), drafts accepted 166 of 218, 1 checkpoints
strata serve: decode expert cache hit rate: 81.4% (105607 hits / 129801 lookups)
strata serve: KV streaming: 99.34% of 1941024 block reads hit VRAM, 51.3 MiB read from RAM
strata serve: suffix drafts: 5 windows, 14 of 15 drafts accepted
