can it run
No. 16GB of memory can't hold the 178GB working set this model needs. A smaller quant or a bigger machine is required.
weights 142GB + KV cache 0.73GB at 4k context vs ~16GB usable → no. 142.7 GB vs 16 GB usable — does not fit
computed roofline: 900 GB/s × 0.46–0.76 efficiency window / 13.3GB (Q4_K_M) — bands, never points.
MoE: file is 142GB (that decides fit) but only ~22B of experts activate per token (~13.3GB streamed) — band below is the active-expert estimate, optimistic; community MoE rows run several× the naive bandwidth math
| model file | 142GB (GGUF Q4_K_M) |
| minimum memory | 178GB working set |
| this machine | 16GB HBM2, 900 GB/s |
| params | 235B-A22B MoE |
| license | apache-2.0 |
| price | $340 |
easy run: ollama pull qwen3:235b-a22b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ ASUS ExpertCenter Pro ET900N G3 (GB300) — cheapest machine that runs it ($60000)
→ every model the Tesla V100 (16GB, used) can run — full list
→ RTX 3090 (24GB, used) — upgrade path machine