AMD Instinct MI350X Can Run Llama 3.1 405B

The AMD Instinct MI350X packs 288GB VRAM. Llama 3.1 405B needs ~222.8GB. You're good to go.

✅ Plenty of headroom

Hardware Specs

GPU
AMD Instinct MI350X
VRAM
288 GB
Bandwidth
6000 GB/s
MSRP
$15,000

Model Requirements

Model
Llama 3.1 405B
Parameters
405B
VRAM (Q4)
~222.8 GB
Context
N/A

Estimated Performance

22 tok/s
Estimated inference speed (Q4 quantization)
Buy vs Rent — Do the Math
Rent breaks even after 51,724hrs • Used pays off faster