Acemagic F9A Mini Workstation Debuts with Ryzen AI Max+ PRO 495 and 192 GB Memory for Local AI

2 Sources

Share

Acemagic unveils its F9A mini workstation at IFA 2026, featuring AMD's Ryzen AI Max+ PRO 495 processor with up to 192 GB unified memory. The compact AI-focused system runs large language models locally, including the 284-billion-parameter DeepSeek V4 Flash, while fitting into a two-liter chassis.

Acemagic F9A Brings Professional-Grade AI Processing to Desktop

Acemagic will showcase its F9A AI mini workstation at IFA 2026 in Berlin from September 4 through September 8, marking one of the first compact systems built around AMD's professional Ryzen AI Max+ platform

1

. The Acemagic F9A measures just 158.5 × 158.5 × 81.5 mm, delivering enterprise-level AI capabilities in a chassis with approximately two liters of volume

2

. This compact AI-focused system positions itself as a practical solution for developers and researchers who need to run substantial AI models locally without relying on cloud infrastructure.

Source: Wccftech

Source: Wccftech

Ryzen AI Max+ PRO 495 Powers Unprecedented Memory Configuration

The flagship configuration pairs the Ryzen AI Max+ PRO 495 processor with up to 192 GB LPDDR5X memory operating at 8533 MT/s speeds. This 16-core, 32-thread processor reaches a maximum boost frequency of 5.2 GHz and integrates Radeon 8065S graphics with 40 RDNA 3.5 compute units running at 3000 MHz

2

. AMD rates the dedicated neural processing unit at 55 TOPS, while combined CPU, GPU, and NPU performance reaches up to 131 TOPS

1

. The unified memory architecture allows the CPU, GPU, and NPU to access the same memory pool, enabling the integrated Radeon 8065S graphics to work with datasets far exceeding the capacity of most discrete graphics cards.

Local AI Model Support Reaches New Scale

Acemagic demonstrated the F9A running the 284-billion-parameter DeepSeek V4 Flash model entirely on local hardware, though the company has not published quantization format, context length, or token-generation speed details

1

. The Ryzen AI Max+ PRO 495 configuration supports up to 120B Q4 models and up to 300B MoE models, representing approximately 60% better AI model support compared to the Ryzen AI Max+ 395 variant

2

. This capability matters because it enables AI-capable mini workstations to handle models that would otherwise require expensive cloud computing resources or high-end workstations with multiple discrete GPUs.

Connectivity Options Support Flexible Workflows

The mini workstation includes two USB4 ports, dual 2.5GbE networking, Wi-Fi 7, Bluetooth 5.4, HDMI 2.1, and DisplayPort 2.1

1

. An OCuLink connection using four PCIe 4.0 lanes provides an expansion path for external graphics or storage hardware with reduced protocol overhead compared to USB4. Internal storage relies on one M.2 2280 PCIe 4.0 x4 slot supporting NVMe SSDs up to 4 TB capacity. These specifications position the F9A for local AI tasks requiring both substantial compute power and flexible data transfer options.

Two Configurations Address Different Performance Tiers

Acemagic will offer the F9A in two variants. The Ryzen AI Max+ 395 model features 16 cores reaching 5.1 GHz boost clock, Radeon 8060S graphics, and up to 128 GB of LPDDR5X memory operating at 8000 MT/s

2

. This configuration supports 70B Q4 models and up to 200B MoE models. Pre-sales for the Ryzen AI Max+ 395 version begin on September 10 through Acemagic's website, with expected pricing between $2999-$3999 based on similarly specified platforms

2

. The Ryzen AI Max+ PRO 495 model remains in preview status without announced pricing, though industry observers anticipate costs exceeding $5000 given the premium memory configuration.

Market Timing Aligns with Rising Demand for Edge AI

The F9A arrives as organizations increasingly prioritize running AI models on local hardware rather than cloud services. This shift stems from concerns about data privacy, API costs, and latency requirements for real-time applications. Mini workstations like the F9A address these needs by delivering substantial compute in compact form factors suitable for office environments. Watch for independent benchmarks measuring actual token-generation speeds, power consumption under sustained AI workloads, and thermal performance in the constrained chassis. The success of this platform will depend on whether the 192 GB LPDDR5X memory configuration justifies its premium pricing for developers working with large language models and multimodal AI systems.

Today's Top Stories

© 2026 TheOutpost.AI All rights reserved