local intelligence

can it run

Can the RTX 3090 (24GB, used) run llava 13B?

YES

Comfortably. 24GB of memory against a 10GB working set leaves headroom for context, the OS, and anything else you're running.

~54–89 tok/s

weights 8GB + KV cache 3.12GB at 4k context vs ~24GB usable → full. 11.1 GB needed of 24 GB usable — headroom for context

computed roofline: 936 GB/s × 0.46–0.76 efficiency window / 8.0GB (Q4_K_M) — bands, never points.

Community check: ~49 tok/s · 12-15B Q4 · 34 runs · inside our computed band — localmaxxing.com

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

Specs

model file8.0GB (GGUF Q4_K_M)
minimum memory10GB working set
this machine24GB GDDR6X, 936 GB/s
params13B
licenseapache-2.0
price$1100

Run it

model files: hugging face ↗

easy run: ollama pull llava:13b

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

Alternatives

no alternatives — every machine in the catalog can run this one.

rtx-5090 — the next step up from this machine's GPU

Buy

NVIDIA (used market) ↗ · Amazon ↗

2026-09-17 · ← full hardware catalog · prices verified at source, may drift