TokenAssemble

Can a Framework Desktop Ryzen AI Max+ 395 128GB run Qwen3-8B?

Yes. Qwen3-8B needs 10.4GB at Q8_0 with llama.cpp and 8K context; the Framework Desktop Ryzen AI Max+ 395 128GB has 128GB of unified memory (~75% GPU-addressable) (95.5GB usable) — grade A+, with decode speed in the usable tier.

Framework Desktop Ryzen AI Max+ 395 128GB · Qwen3-8B · Q8_0

Fits — grade A+

UsablebetaROCm/Vulkan performance not yet calibrated
weights 8.71 GBKV cache 1.21 GBoverhead 0.48 GBpool 95.5 GBneeds 10.4 GB

assumes llama.cpp · batch 1 · 8K context · fp16 KV cache

$3,449 MSRP · as of 2026-07-11

Affiliate link — we may earn a commission. Verdicts are computed before any link is attached.

Every quant, graded

Framework Desktop Ryzen AI Max+ 395 128GB × Qwen3-8B, llama.cpp, 8K context.

QuantFile sizeFitSpeed tier
Q8_08.7 GBFits · A+usable · beta
Q6_K6.7 GBFits · A+interactive · beta
Q5_K_M5.8 GBFits · A+interactive · beta
Q4_K_M5.0 GBFits · A+interactive · beta

Qwen3-8B on similar hardware

More models on the Framework Desktop Ryzen AI Max+ 395 128GB

Verdict computed from published specs and measured GGUF file sizes — see how the math works.