strata serve: prompt 4022 tokens = 0 reused + 4022 read in 12723 ms (316.1 tok/s), 256 generated in 5398 ms (47.4 tok/s), drafts accepted 165 of 218, 1 checkpoints
strata serve: decode expert cache hit rate: 81.5% (106088 hits / 130162 lookups)
strata serve: KV streaming: 99.35% of 1947180 block reads hit VRAM, 51.4 MiB read from RAM
strata serve: suffix drafts: 5 windows, 14 of 15 drafts accepted
