local intelligence

can it run

Can the RTX 2060 12GB (used) run qwen3-coder 480B A35B cloud-class?

NO

No. 12GB of memory can't hold the 363GB working set this model needs. A smaller quant or a bigger machine is required.

~0.56–0.83 tok/s

weights 290GB + KV cache 0.97GB at 4k context vs ~6GB usable → no. 291.0 GB vs 6 GB usable — does not fit

computed roofline: 336 GB/s × 0.48–0.72 efficiency window / 290.0GB (Q4_K_M) — bands, never points.

explore the full catalog84 machines indexed · live prices · what each one can run

Specs

model file290GB (GGUF Q4_K_M)
minimum memory363GB working set
this machine12GB GDDR6, 336 GB/s
params480B A35B cloud-class
licenseapache-2.0
price$375

Run it

model files: hugging face ↗

easy run: ollama pull qwen3-coder:480b

runs with: ollama · llama.cpp · koboldcpp · lm studio

Alternatives

ASUS ExpertCenter Pro ET900N G3 (GB300) — cheapest machine that runs it ($60000)

every model the RTX 2060 12GB (used) can run — full list

rtx-5060-ti-16 — the next step up from this machine's GPU

Buy

eBay (used) ↗ · Amazon ↗

2026-09-17 · ← all models on the RTX 2060 12GB (used) · ← full hardware catalog · prices verified at source, may drift