ComfyICU / infrastructure evidence
Local GPU infrastructure
Only fixed-protocol runs: filesystem pages evicted before every cold process, physical reads verified, then a different-seed warm run in the same process.
Dynamic: DuckDB queries completed JSON files on every request. New verified runs appear without regenerating this page. Browse docs →
Storage path vs cold penalty
model-file GiB/s → cold − warm seconds| Run | Storage | Cold | Warm | Penalty | Model read | Comfy read | fio QD1 / QD32 | PCIe H2D | Evidence |
|---|---|---|---|---|---|---|---|---|---|
| Loading DuckDB results… | |||||||||
Measured economics / all GPU classes
Rental and purchase Pareto
Every performance point comes from a successful five-pair schema-v3 run. Rental uses the Vast price captured with that exact host. Purchase prices are explicit, editable inputs joined to the same measured throughput.
Normalize the quote scope before buying. Some inputs are card-only marketplace asks while “White Box” supplier quotes may include more of a system. Capex rankings are provisional until every quote uses the same bill of materials.
Rental price vs measured throughput
lower cost → / higher throughput ↑| Loading economics… |