local intelligence

can it run

Can the GTX 970M run WHISPER base?

FULL

2.3 GB needed of 3 GB usable — headroom for context

3GB · Maxwell (laptop) · model file 0.3GB · base

~110–200 tok/s

computed roofline: 120.3 GB/s × 0.28–0.50 efficiency window / 0.3GB (Q4_K_M) — bands, never points.

Run WHISPER base on instead

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

every model the GTX 970M can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the GTX 970M · fit = weights + 4k KV vs usable memory