AMD Instinct MI250X~515 tok/s (Q4)

ibm-granite/granite-3.3-8b-instruct speed on AMD Instinct MI250X

Quantization-specific throughput and VRAM requirements for ibm-granite/granite-3.3-8b-instruct running on AMD Instinct MI250X.

Speed Snapshot

Topline estimate from compatibility data

Modelibm-granite/granite-3.3-8b-instruct

GPUAMD Instinct MI250X

Q4 speed515 tok/s

Q4 VRAM required4GB

Data Source

Calculation and benchmark status

Speed values come from the compatibility dataset (`estimatedTokensPerSec`) and are sorted by quantization.

For full verdict logic and alternate GPUs, see the canonical compatibility page.

Quantization Speed Table

Quantization	VRAM needed	VRAM available	Speed	Verdict
Q4	4GB	128GB	515 tok/s	✅ Fits
Q8	9GB	128GB	333 tok/s	✅ Fits
FP16	17GB	128GB	182 tok/s	✅ Fits