Hardware Detail AI PCPrevious gen

Ryzen AI 9 HX 370 (Strix Halo)

Ryzen AI 9 HX 370 (Strix Halo) profile for local AI model planning.

AMD (ASUS/Lenovo/HP)AMD AI PC2024-1032GB unified

AI Memory

32GB unified

Usable model memory reserves headroom for OS, runtime, and KV cache.

LLM Score

prosumer

Power

45W

Average AI load

Value Score

7.5

Planner signal

Task Fit

Chat72

Usable for chat, with tradeoffs around speed, memory, or runtime maturity.

Code70

Usable for code, with tradeoffs around speed, memory, or runtime maturity.

RAG70

Usable for rag, with tradeoffs around speed, memory, or runtime maturity.

Agent67

Usable for agent, with tradeoffs around speed, memory, or runtime maturity.

Vision62

Usable for vision, with tradeoffs around speed, memory, or runtime maturity.

Embedding56

Not a primary fit for embedding in the current hardware pool.

Voice53

Not a primary fit for voice in the current hardware pool.

Image generation41

Not a primary fit for image generation in the current hardware pool.

Video generation28

Not a primary fit for video generation in the current hardware pool.

Runtime Support

ollamaPartial

llama.cppgood

LM StudioPartial

vllmPartial

sglangPartial

ComfyUIgood

diffusersPartial

Specs

CPU cores12

GPU cores—

System RAM32GB

Unified memory32GB

VRAM—

Usable model memory24GB

CUDANo

MetalNo

ROCmSupported

Cost and Ops

Hardware price$1,499

Build cost—

UPS / accessories$150

5-year estimate$1,647

Expected lifespan5 years

Residual value25%

Build requiredNo

Models This Hardware Can Run

Estimated from default model artifacts, VRAM, unified memory, and recommended RAM.

14 compatible defaults

Gemma 4 12B It QAT W4a16 Ct

Compatible

Google · Q4_0

16GB recommended

Gemma 4 E4B It QAT Q4 0 GGUF

Compatible

Google · Q4_0

12GB recommended

Kimi VL A3B Thinking 2506

Compatible

Moonshot AI · Q4_K_M

8GB recommended

Gemma 4 E2B It QAT Mobile Transformers

Google · Q4_K_M

Alibaba · Q4_K_M

· FP16

Nomic Embed Text v1.5

· FP16

OpenAI · FP16

· FP16

Best For

Windows AI PCbalance of CPU/GPU/NPURyzen AI SDK

Ryzen AI 9 HX 370 (Strix Halo) profile for local AI model planning.

Source Confidence

SpecsEstimated

PriceEstimated

PowerEstimated

PerformanceEstimated

Updated2026-06-18

Similar Hardware

Ryzen AI Max+ 395 128GB Mini PC

AMD / Mini PC OEMs · 128GB unified · $2,311 / 5y

Ryzen AI Max+ 395 128GB Mini PC profile for local AI model planning.

Minisforum UM890 (Ryzen 9 8945HS)

Minisforum · 32GB unified · $597 / 5y

Minisforum UM890 (Ryzen 9 8945HS) profile for local AI model planning.

Framework Desktop AI Max 128GB

Framework / AMD · 128GB unified · $2,360 / 5y

Framework Desktop AI Max 128GB profile for local AI model planning.