local intelligence

can it run

Can the NVIDIA L4 run PHI4 mini 3.8B?

FULL

3.0 GB needed of 24 GB usable — headroom for context

24GB · Ampere · model file 2.5GB · mini 3.8B

~66–110 tok/s

computed roofline: 300 GB/s × 0.55–0.88 efficiency window / 2.5GB (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 PHI4 mini 3.8B on instead

Raspberry Pi 5 + AI Kit (Hailo-8L) — cheapest catalog machine that runs PHI4 mini 3.8B fully ($110)

RTX 4090 (24GB, used) — upgrade path machine

every model the NVIDIA L4 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 L4 · fit = weights + 4k KV vs usable memory