GMK Unveils 192GB AI Mini PC
💡A 192GB unified-memory mini PC could run 300B AI models locally without a discrete GPU.
⚡ 30-Second TL;DR
What Changed
Powered by a 16-core, 32-thread AMD Ryzen AI Max+ PRO 495 chip based on Zen 5.
Why It Matters
This could make high-parameter local inference more accessible without a discrete GPU or cloud deployment. However, practical performance will depend on model quantization, memory bandwidth utilization, software support, and sustained thermal limits.
What To Do Next
Benchmark a quantized 70B-to-300B model with llama.cpp or vLLM on the EVO-X5 Pro before committing to local inference deployment.
Key Points
- •Powered by a 16-core, 32-thread AMD Ryzen AI Max+ PRO 495 chip based on Zen 5.
- •Includes 192GB LPDDR5X-8533 unified memory with 273GB/s bandwidth.
- •The Radeon 8065S iGPU can access up to 160GB and the system supports local 300B AI models.
- •Connectivity includes four USB4-class ports, dual 10GbE networking, and a configurable TDP button.
- •GMK has not announced pricing and currently plans only the flagship 192GB configuration.
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: IT之家 ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
Weekly AI briefing
One email a week. Unsubscribe anytime.