local intelligence

can it run

Can the GTX 980M run mistral 7B?

OFFLOAD PARTIAL

4.6 GB vs 4 GB usable — partial CPU offload, expect large speed loss With a q8 (quantized) KV cache it comes to the edge (4.3GB) — possible with modest context, at some quality cost.

4GB · Maxwell (laptop) · model file 4.1GB · 7B

~11–20 tok/s

computed roofline: 160.4 GB/s × 0.28–0.50 efficiency window / 4.1GB (Q4_K_M) — bands, never points.

Run mistral 7B on instead

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

every model the GTX 980M 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 980M · fit = weights + 4k KV vs usable memory