can it run
No. 16GB of memory can't hold the 22GB working set this model needs. A smaller quant or a bigger machine is required.
weights 17GB + KV cache 2GB at 4k context vs ~16GB usable → offload-partial. 19.0 GB vs 16 GB usable — partial CPU offload, expect large speed loss
computed roofline: 160.4 GB/s × 0.28–0.50 efficiency window / 17.0GB (Q4_K_M) — bands, never points.
| model file | 17GB (GGUF Q4_K_M) |
| minimum memory | 22GB working set |
| this machine | 16GB GDDR5 (2×8 GB), 160 GB/s |
| params | schnell fp8 |
| license | flux-dev-nc / apache schnell |
| price | $35 |
easy run: no stock runtime yet
→ Tesla V100 32GB (used) — cheapest machine that runs it ($645)
→ every model the Tesla M60 (16GB, used) can run — full list
→ tesla-p100-16gb — the next step up from this machine's GPU