What can a Ryzen AI Max+ 395 128GB run?

128GB · 256 GB/s · 120W · $2500

AMD's 128 GB APU monster — desktop-class unified memory for big models at a fractal power draw.

Runs great22

Comfortable headroom — daily-driver territory.

Model	Quant	Est. speed	Max context
deepseek moe 16b chat	FP16	Infinity tok/s	—
qwen2 7b instruct	FP16	Infinity tok/s	—
Qwen3-0.6B	FP16	128 tok/s	33k
Gemma-3n-E4B	FP16	57 tok/s	131k
GPT-OSS-120B	Q4_K_M	51 tok/s	103k
SmolLM2-1.7B	FP16	47 tok/s	8k
Llama-3.2-3B	FP16	25 tok/s	131k
Qwen3-30B-A3B	FP16	24 tok/s	33k
Qwen3-Coder-30B-A3B	FP16	24 tok/s	195k
GPT-OSS-20B	FP16	22 tok/s	131k
Mistral-7B-Instruct	Q8_0	21 tok/s	33k
Phi-4-mini	FP16	21 tok/s	131k
Gemma-3-4B	FP16	21 tok/s	131k
Qwen3-4B	FP16	20 tok/s	33k
DeepSeek-R1-Distill-Qwen-7B	Q8_0	20 tok/s	131k
Qwen2.5-Coder-7B	Q8_0	20 tok/s	131k
Llama-3.1-8B	Q8_0	19 tok/s	131k
Qwen3-8B	Q8_0	19 tok/s	33k
Gemma-3-12B	Q5_K_M	17 tok/s	131k
Qwen3-14B	Q4_K_M	16 tok/s	33k
Phi-4-14B	Q5_K_M	15 tok/s	16k
DeepSeek-R1-Distill-Qwen-14B	Q5_K_M	15 tok/s	131k

Tight squeeze14

It fits, but close the browser tabs and mind the context.

Model	Quant	Est. speed	Max context
Qwen3.8 Flash Next	Q3_K_M	108 tok/s	24k
Qwen3-235B-A22B	IQ2_M	21 tok/s	33k
Codestral-22B	FP16	4 tok/s	33k
Mistral-Small-24B	FP16	3 tok/s	131k
Gemma-3-27B	FP16	3 tok/s	131k
qwen35 27B	FP16	3 tok/s	241k
Qwen3.8	FP16	3 tok/s	246k
Qwen3.8 27B	FP16	3 tok/s	246k
DeepSeek-R1-Distill-Qwen-32B	FP16	2 tok/s	131k
GLM-4-32B	FP16	2 tok/s	131k
Qwen3-32B	FP16	2 tok/s	33k
Llama-3.3-70B	Q8_0	2 tok/s	89k
Qwen2.5-72B	Q8_0	2 tok/s	77k
glm-4.6-air 106B	Q6_K	2 tok/s	20k

Measured on this card0

Everything here was benchmarked on this exact card by someone who owns one — the tables above are our arithmetic; these numbers came off real machines.

Nobody has measured anything on this card yet. Paste a benchmark — it takes one keystroke and no form.

Try your own machine →← All hardware