AMD Instinct MI250X~605 tok/s (Q4)

google/embeddinggemma-300m speed on AMD Instinct MI250X

Quantization-specific throughput and VRAM requirements for google/embeddinggemma-300m running on AMD Instinct MI250X.

Speed Snapshot

Topline estimate from compatibility data

Modelgoogle/embeddinggemma-300m

GPUAMD Instinct MI250X

Q4 speed605 tok/s

Q4 VRAM required1GB

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	1GB	128GB	605 tok/s	✅ Fits
Q8	1GB	128GB	428 tok/s	✅ Fits
FP16	1GB	128GB	236 tok/s	✅ Fits