strata serve: prompt 94 tokens = 0 reused + 94 read in 2995 ms (31.4 tok/s), 256 generated in 6259 ms (40.9 tok/s), drafts accepted 156 of 221, 1 checkpoints
strata serve: decode expert cache hit rate: 79.8% (106086 hits / 132985 lookups)
strata serve: KV streaming: 99.51% of 217332 block reads hit VRAM, 4.3 MiB read from RAM
