local intelligence

can it run

Can the NVIDIA GeForce RTX 2070 Super run llava 7B?

TIGHT

6.7 of 8 GB usable — keep context modest

8GB · Turing TU104 · model file 4.7GB · 7B

~46–69 tok/s

computed roofline: 448 GB/s × 0.48–0.72 efficiency window / 4.7GB (Q4_K_M) — bands, never points.

Run llava 7B on instead

Raspberry Pi 5 + AI Kit (Hailo-8L) — cheapest catalog machine that runs llava 7B fully ($110)

rtx-3060-12 — the next step up from NVIDIA GeForce RTX 2070 Super

every model the NVIDIA GeForce RTX 2070 Super can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the NVIDIA GeForce RTX 2070 Super · fit = weights + 4k KV vs usable memory