Mini PC Minis Forum MS-S1 Max AMD Ryzen AI Max+ 395 128GB Radeon Graphics

MinisForum MS-S1 Max

Your advanced partner for every AI task

For AI enthusiasts

Perfect for running large models locally, developing AI applications, and real-time experimentation—with minimal latency and consistent performance. Whether it’s model optimization, inference, or fine-tuning, the MS‑S1 Max provides an efficient workspace without cloud dependence.

For power users

Enables rapid testing of new AI frameworks, local deployment of models, and running heavy workloads thanks to fast multi-channel memory and advanced graphical and neural compute power. Ideal for a development environment with multiple tools and containers.

For professional creators

Significantly accelerates video editing, 3D rendering, AI-based art creation, and audio production. Up to 128GB of UMA memory and multi-display support enable a smooth workflow even in complex projects, with fast and stable results.

A compact supercomputer for the next AI generation

Powered by AMD Ryzen AI Max+ 395, with Zen 5 cores, RDNA 3.5 graphics, and an advanced dedicated NPU. This combination offers an excellent performance-to-power ratio—ideal for always-on AI.

16 cores and 32 threads, GPU performance comparable to an RTX 4070 Laptop, and 50 TOPS on the NPU for fast, efficient AI acceleration.

Reaches up to 126 TOPS total, with AI performance up to 2.2× that of an RTX 4090 in selected scenarios—suited for running LLMs, multimodal creation, computer vision, and more.

Supports LPDDR5x memory at 8000 MT/s up to 128GB in a four-channel configuration for ultra-fast data transfer and instant responsiveness under load.

From desktop to compute node—building a smart infrastructure

Two MS‑S1 Max devices can be linked in parallel (cluster) to scale computing power as needed—for personal research, studios, and small teams. Thanks to the modular design, expansion is simple and user-friendly.

In practical tests, connecting two devices runs a 235B Q4 model locally at an output speed of 10.87 tokens per second—an impressive result for an on‑prem environment.

Supports installation in a 2U chassis, with the option to connect multiple units to create a distributed cluster—an efficient solution for AI labs, R&D teams, and real-time AI applications.

Advanced cooling for consistent performance

Aluminum chassis with a diamond-pattern grill for optimal ventilation, combining impressive durability with modern design. The engineered airflow maintains stable operating temperatures over time.

Dual turbine fans create focused airflow for efficient heat dissipation, reduced noise, and prevention of throttling even under heavy, sustained loads.

Secure and efficient internal power supply

320W internal power supply with CCC, FCC, UL, CE, UKCA, and more certifications—for stable, safe, and quiet power delivery. Built for prolonged operation with full voltage protections.

Upgrades and maintenance—made simple

A removable design with an integrated board tray provides quick access to components, allowing easy upgrades, replacements, and maintenance—saving time and ensuring high system availability.

Storage and expansion

Two M.2 slots (PCIe 4.0 x4 + x1) with a total capacity of up to 16TB (8TB + 8TB) and RAID 0/1 support. The removable board makes installation and maintenance simple, even in multi-drive setups.

Full-length expansion

Full-length PCIe x16 slot for specialized needs—custom I/O interfaces, accelerator cards, communication modules, and more. Maximum flexibility for system customization.

Comprehensive connectivity

1. Power button

2. Pair of digital microphones

3. USB 3.2 Gen2 (10Gbps)

4. USB4 ×2 (40Gbps, Alt DP 2.0, 15W PD output)

5. 3.5mm combo audio jack

6. HDMI 2.1 FRL (up to 8K@60Hz / 4K@120Hz)

7. USB 2.0 ×2

8. USB4 V2 ×2 (80Gbps, Alt DP 2.0, 15W PD output)

9. USB 3.2 Gen2 (10Gbps)

10. 10GbE LAN ×2 (RJ45, Realtek RTL8127)

11. Lock slot

12. Reset hole (Clear CMOS)

13. Power input (AC INPUT)