ProductAMD AI Max Plus 395

AMD AI Max+ 395

A high-performance AI acceleration processor from AMD featuring a unified memory architecture. It provides large memory capacity and high memory bandwidth, making it well suited for the Decode stage of large language model inference workloads.

Timeline (last 90 days)

Sep 26

In Path A, an RTX 3060 runs the front layers and an AMD AI Max+ 395 runs the back layers, improving Prefill speed from 136.69 token/s to 319.1 token/s under IQ3 quantization with 64K long input.

Unverified50%

All Facts (1)

Unverified

In Path A, an RTX 3060 runs the front layers and an AMD AI Max+ 395 runs the back layers, improving Prefill speed from 136.69 token/s to 319.1 token/s under IQ3 quantization with 64K long input.

50%

Source Articles

Deploying Qwen3 on AMD AI MAX+ 395: GFX1151 Engine Optimization Deep DiveSep 26 GPU Prices Going Crazy? Deploy LLMs with Unified Memory + Discrete GPU Heterogeneous AccelerationSep 26