local intelligence

can it run

Can the NVIDIA GeForce RTX 2080 Super run gemma3 1B?

FULL

1.1 GB needed of 8 GB usable — headroom for context

8GB · Turing TU104 · model file 1.0GB · 1B

~240–360 tok/s

computed roofline: 495.9 GB/s × 0.48–0.72 efficiency window / 1.0GB (Q4_K_M) — bands, never points. Small models (under ~4.5B) run BELOW these bands — kernel-launch overhead keeps them from saturating bandwidth (measured against community medians, 09-06).

Run gemma3 1B on instead

Raspberry Pi 5 + AI Kit (Hailo-8L) — cheapest catalog machine that runs gemma3 1B fully ($110)

rtx-3080-10gb — the next step up from NVIDIA GeForce RTX 2080 Super

every model the NVIDIA GeForce RTX 2080 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 2080 Super · fit = weights + 4k KV vs usable memory