← AI Hashrate 中文

Ryzen AI Max 390 64GB vs RTX 5090 32GB for Local LLMs

The RTX 5090 32GB has about 733% more memory bandwidth (1,792 vs 215 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 733% faster on the RTX 5090 32GB. The Ryzen AI Max 390 64GB fits more of the catalog at Q4/8K — 41 vs 39 of 69 models (48 vs 32 GB VRAM). Largest model that fits the Ryzen AI Max 390 64GB but not the RTX 5090 32GB: Qwen3-Next-80B-A3B-Instruct (81.3B). MSRP is $1,300 for the Ryzen AI Max 390 64GB vs $1,999 for the RTX 5090 32GB. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. In the Q4/8K head-to-head below (53 shared models), the RTX 5090 32GB posts the higher tok/s in 52 rows vs 1. Methodology.

Spec comparison

Ryzen AI Max 390 64GBRTX 5090 32GB
VendorAMDNVIDIA
VRAM48 GB32 GB
Memory typeLPDDR5XGDDR7
Memory bandwidth215 GB/s1,792 GB/s
TDP120 W575 W
MSRP (list)$1,300$1,999
BuySearch Amazon (affiliate)Search Amazon (affiliate)

MSRP is the launch list price, not live retail. Buy links are Amazon affiliate links — the price after you click is the live price.

Q4 / 8K head-to-head — top 15 of 53 shared models

ModelParams (B)Ryzen AI Max 390 64GB tok/sRyzen AI Max 390 64GB fitsRTX 5090 32GB tok/sRTX 5090 32GB fitsWinner
MiniCPM5-1B1.08126.7 est.Yes1055.9 est.YesRTX 5090 32GB
Qwen3.5-2B2.068.4 est.Yes570.2 est.YesRTX 5090 32GB
Qwen3-Coder-Next79.745.6 est.Yes550.5 est.NoRTX 5090 32GB
Qwen3-Next-80B-A3B-Instruct81.345.6 est.Yes529.6 est.NoRTX 5090 32GB
DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx2.359.5 est.Yes495.8 est.YesRTX 5090 32GB
Qwen3.6-35B-A3B36.045.6 est.Yes380.1 est.YesRTX 5090 32GB
Qwen3.5-35B-A3B34.745.6 est.Yes380.1 est.YesRTX 5090 32GB
GLM-4.7-Flash31.245.6 est.Yes380.1 est.YesRTX 5090 32GB
Qwen3-30B-A3B-Instruct-250730.545.6 est.Yes380.1 est.YesRTX 5090 32GB
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF1630.045.6 est.Yes380.1 est.YesRTX 5090 32GB
Ornith-1.0-35B35.045.6 est.Yes380.1 est.YesRTX 5090 32GB
qwen35b-a3b-fable-sft-abliterated34.745.6 est.Yes380.1 est.YesRTX 5090 32GB
Llama-3.2-3B-Instruct3.242.8 est.Yes356.4 est.YesRTX 5090 32GB
gemma-4-26B-A4B-it25.236.0 est.Yes300.1 est.YesRTX 5090 32GB
Qwen3-4B-Instruct-25074.034.2 est.Yes285.1 est.YesRTX 5090 32GB

est. = bandwidth-based estimate; measured = curated real benchmark that overrides the estimate. Fits = weights + KV(8K) + 1 GB overhead ≤ 95% of VRAM.

Fit coverage at Q4 / 8K

Ryzen AI Max 390 64GBRTX 5090 32GB
Catalog models that fit41 of 6939 of 69
Largest exclusive fitQwen3-Next-80B-A3B-Instruct (81.3B)

Ryzen AI Max 390 64GB details → · RTX 5090 32GB details →