The Definitive Guide toAI Data Centers
Ask the GuideAboutAccount
GuideGlossaryMIG

MIG · Multi-Instance GPU

An NVIDIA feature partitioning one GPU into isolated instances so multiple workloads share it securely.

Current numbers

~2/3inference share of AI compute in 2026 — the workload class that most rewards fractional/MIG sharingas of 2026 · register ↗
up to 7MIG instances per GPU (B200/GB200): 2×~93GB, 4×~46GB, or 7×~23GB profilesas of 2025 · register ↗
7max MIG instances per GPU — the only hardware-enforced fractional partition (dedicated SMs, L2 slice, memory controllers, HBM slice)as of 2025 · register ↗

← All terms