Free tools Windows power users keep installed
One-click scans. No signup required.
The Ryzen AI Max+ 395 is a platform, not a guarantee that every mini PC performs the same. AMD’s chip combines 16 Zen 5 CPU cores with Radeon 8060S integrated graphics and supports up to 128GB of fast unified memory. In practice, the system’s installed memory, soldered design, cooling, power limits, storage expansion and exact regional SKU matter as much as the processor name. Framework offers documented 64GB and 128GB configurations; HP’s cited Z2 Mini G1a configuration has 32GB onboard; and a secondary report describes the MOGINSOK MG01 with 128GB. Compare those complete systems rather than assuming identical performance from the shared 395 designation.
What the Ryzen AI Max+ 395 platform includes
AMD specifies the Ryzen AI Max+ 395 as a 16-core, 32-thread Zen 5 processor with integrated Radeon 8060S graphics. The graphics implementation contains 40 graphics cores and uses the same unified memory pool as the CPU. AMD’s platform specification allows up to 128GB of 256-bit LPDDR5x memory at speeds up to 8000 MT/s. Those are platform limits, not promises that every finished computer will ship with that capacity or speed.
As an Amazon Associate I earn from qualifying purchases.
Unified memory is important for this class of machine. The processor and graphics engine draw from the same installed pool, so a 32GB system has less headroom for large applications, high-resolution textures or local AI models than a 128GB system. Allocating more memory to graphics can also leave less available to the operating system and applications. Because the memory is soldered in the documented Framework and HP examples, capacity is generally a purchase-time decision rather than an upgrade you can make later.
At-a-glance comparison
| System | Processor and graphics | Installed memory | Storage and expansion | Form factor and notable details |
|---|---|---|---|---|
| Framework Desktop (2025 configuration) | Ryzen AI Max+ 395; Radeon 8060S, 40 compute units, up to 2.9GHz | 64GB or 128GB soldered LPDDR5x-8000 | Two M.2 2280 PCIe 4.0 x4 sockets | 4.5-liter desktop; supports up to four displays; processor power listed up to 120W sustained and 140W boost |
| HP Z2 Mini G1a, cited E07RCPA configuration | Ryzen AI Max+ PRO 395; Radeon 8060S | 32GB onboard LPDDR5x-8533 | 1TB PCIe Gen4 NVMe SSD | Regional HP specification; verify the exact country and SKU |
| MOGINSOK MG01, configuration reported October 2026 | Ryzen AI Max+ 395 | 128GB LPDDR5x | 1TB SSD; report mentions four M.2 slots and dual networking | Secondary retail report; current seller, warranty and exact configuration require verification |
| ASUS ROG Flow Z13 (2025) | Ryzen AI Max+ 395; Radeon 8060S | 128GB onboard LPDDR5X-8000 in the cited listing | 1TB M.2 2230 SSD | 13.4-inch 2.5K 180Hz convertible tablet/laptop, not a mini PC |
The Framework and HP specifications are manufacturer-published. The MG01 details come from a TechRadar report that linked to an Amazon offer, so treat them as a product lead until the live listing, seller, SKU and support terms are checked. The Flow Z13 is useful as a portability comparison, but its battery, screen and convertible chassis make it a different buying decision from a desktop mini PC.
#1 Best Overall
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 64GB pool, which is perfect for running LLMs such as Deepseek 32B, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 4% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
Which Ryzen AI Max+ mini PC has 128GB of memory?
Three entries in the cited material reach 128GB, but they are not equivalent choices. Framework’s 2025 Desktop configuration offers a documented 128GB soldered option alongside 64GB. The reported MOGINSOK MG01 configuration also lists 128GB, although its current availability and specifications need confirmation. ASUS lists 128GB onboard in the 2025 ROG Flow Z13, which is a portable convertible rather than a mini PC.
HP’s cited Z2 Mini G1a E07RCPA configuration has 32GB onboard, despite using the same general 395-class silicon. That lower capacity may be adequate for office work, development and many creative applications, but it leaves less room for memory-heavy workloads than a 64GB or 128GB machine. HP configurations can vary by region, so do not generalize the 32GB figure to every Z2 Mini G1a listing.
Memory capacity, speed and upgradeability
Capacity determines what fits
For ordinary productivity, 32GB can be comfortable. Developers running containers and virtual machines, creators working with large media projects, and users loading local language or vision models benefit from 64GB or more. A 128GB system is the safer choice when the computer must keep a large model, several applications and substantial graphics allocations in memory at the same time.
Speed is not a complete performance ranking
The HP example lists LPDDR5x-8533, while Framework lists LPDDR5x-8000. A higher memory-transfer figure does not by itself prove that the HP system is faster overall. CPU and GPU power limits, cooling, firmware, memory timings, storage and workload behavior all affect results. No standardized independent test in the cited material isolates those memory-speed differences across these mini PCs.
Rank #2
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 128GB pool, which is perfect for running LLMs such as Deepseek 70B Q8, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 12% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
Soldered memory changes the buying decision
Framework describes its 395 memory as soldered, and HP calls the cited memory onboard. Confirm whether a specific SKU is soldered before ordering. If it is, select the capacity you expect to need over the computer’s useful life; adding an SSD later is not a substitute for insufficient system memory because storage is dramatically slower and cannot provide the same graphics bandwidth.
How Radeon 8060S graphics should be compared
Radeon 8060S is the common graphics name in the Framework, HP and Flow Z13 specifications, and AMD lists 40 graphics cores for the Ryzen AI Max+ 395 platform. That common label provides a useful baseline, but it does not make the systems interchangeable. Integrated graphics performance depends on the power available to the processor package, cooling capacity, firmware settings and how much unified memory is reserved for graphics.
Framework lists up to 120W sustained and 140W boost processor power for its 2025 Desktop configuration. A smaller or more thermally constrained enclosure may use different limits. Sustained rendering, compiling or gaming loads can therefore produce different behavior even when the CPU and GPU names match. The manufacturer specifications cited here do not provide a common sustained-performance test, so a precise FPS or rendering ranking would be unsupported.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →For gaming, check the resolution, quality preset, frame-rate target and whether the title supports upscaling before treating any 8060S claim as meaningful. For professional graphics, verify application support, display outputs and the memory required by the project. Framework documents support for up to four displays; other systems may expose different combinations of HDMI, DisplayPort, USB-C or dock connectivity.
Rank #3
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 64GB pool, which is perfect for running LLMs such as Deepseek 32B, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 4% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
Can a Ryzen AI Max+ 395 mini PC run local AI models?
Yes, the 395 platform is suitable for local AI experimentation, but the model size and software stack determine whether it is practical. Unified memory lets the CPU and integrated GPU share a large pool, making 64GB and 128GB configurations particularly interesting for models that exceed the comfortable capacity of typical integrated-graphics laptops. A 32GB machine will encounter the limit sooner.
AMD’s March 17, 2025 LM Studio article used an ASUS ROG Flow Z13 with 64GB of unified memory. In AMD’s selected comparison, the company reported up to 2.2× token throughput and, for a selected 14-billion-parameter model, up to 12.2× faster time to first token against an Intel Arc 140V laptop. These are AMD’s own results from a specific laptop, software setup, model selection and comparison hardware. They are not an independent ranking of the mini PCs above and should not be treated as a guaranteed result for every 395 system.
When evaluating local AI, record the model quantization, context length, prompt and generation settings, backend, driver version, memory allocation and sustained power behavior. Report both prompt-processing speed and generated-token speed where possible. A short burst can look impressive while a long session slows as the chassis reaches its thermal limit.
Cooling, power and chassis differences
Mini PCs trade volume for density. Two systems with the same processor can differ in fan noise, surface temperature and sustained clock speeds because their heatsinks, fans and power supplies are different. A 4.5-liter desktop such as the documented Framework configuration has more physical room than an ultra-compact design, while a portable convertible must also balance battery operation and tablet ergonomics.
Rank #4
- Unlock next-generation AI computing with AMD Ryzen AI Max+ 395 processor featuring 16 cores, 32 threads, up to 5.1GHz boost clock, and integrated Ryzen AI engine delivering up to 126 TOPS AI performance. EVO-X3 is designed for local AI models, content creation, development, and professional workloads.
- OCuLink External GPU Expansion – Upgrade Beyond a Mini PC: Take your graphics performance further with a dedicated OCuLink (PCIe 4.0 x4) interface. Connect an external GPU dock to add desktop-class graphics power for AAA gaming, AI acceleration, 3D rendering, video production, and advanced creative applications. EVO-X3 gives you the flexibility of a compact PC with workstation-level expansion capability.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 128GB pool, which is perfect for running LLMs such as Deepseek 70B Q8, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 12% better performance in digital content workloads.
Look for published sustained and boost power limits, then check whether those figures apply to the exact configuration you are buying. Boost power is a short-term ceiling, not a promise that the machine can maintain it during a long encode or compile. Reviews that measure ten-minute or longer workloads are more useful for this question than a single short benchmark run, but the cited material does not provide a standardized cross-model test.
Storage, ports and expansion
Framework Desktop
The 2025 395 configuration documents two M.2 2280 PCIe 4.0 x4 sockets. That gives the system a clear path for adding or separating fast storage, subject to the exact configuration and thermal design. It also lists support for up to four displays, which suits multi-monitor workstations.
HP Z2 Mini G1a
The cited E07RCPA specification includes a 1TB PCIe Gen4 NVMe SSD. The source does not establish the full expansion layout for every regional Z2 Mini G1a SKU, so confirm available M.2 sockets, drive length, service access and included networking before purchase.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →MOGINSOK MG01
The October 2026 report mentions four M.2 slots and dual networking, but it also says additional information for a claimed SATA detail could not be found. Verify the live product page, photographs, manual and seller response rather than assuming a SATA bay or a particular slot type exists.
ROG Flow Z13
The cited ASUS configuration uses a 1TB M.2 2230 SSD. Its compact convertible design prioritizes portability and a built-in 13.4-inch 2.5K 180Hz touchscreen rather than desktop-style drive expansion.
How to choose the right configuration
- Set the memory floor. Choose 32GB only when your workloads are known to be light. Select 64GB for mixed development, creative and AI use, and 128GB when large models, virtual machines or heavy multitasking are central.
- Confirm the exact processor label. Distinguish Ryzen AI Max+ 395 from Ryzen AI Max PRO generations. Framework’s current product pages also describe newer 2026 Ryzen AI Max PRO products; do not compare those directly with the documented 2025 395 configuration without checking the chip.
- Check whether memory is soldered. Treat onboard or soldered capacity as permanent and buy the amount you need at checkout.
- Match expansion to the workload. Count M.2 sockets, drive lengths, display outputs, USB ports and networking interfaces on the exact SKU.
- Evaluate sustained operation. Look for power limits, cooling measurements, fan-noise data and long-running tests rather than relying on the processor name.
- Verify region, seller and warranty. HP’s cited configuration is regional, and the MG01 information is based on a reported retail offer. Confirm stock, currency, return terms, support coverage and the installed memory and SSD before paying.
Common buying mistakes
- Assuming all 395 systems perform identically: power limits and cooling can change sustained behavior.
- Comparing memory speed without capacity: 8533 MT/s does not compensate for having half the memory needed by a workload.
- Calling the Flow Z13 a mini PC: it is a convertible tablet/laptop with a screen and battery.
- Treating AMD’s AI figures as universal: the 2.2× and 12.2× results are vendor-published, selected comparisons.
- Trusting a product-family name: regional SKUs can change memory, storage and connectivity.
- Planning to upgrade soldered memory: choose capacity before purchase and use replaceable SSDs for later expansion.
Bottom line for buyers
Choose the Framework Desktop when documented memory options, two M.2 sockets, multi-display support and a larger desktop chassis are priorities. Consider the HP Z2 Mini G1a when its regional workstation configuration, support terms and 32GB capacity fit your workload. Investigate the MOGINSOK MG01 only after verifying its current 128GB listing, four-slot claim, seller and warranty. Select the ROG Flow Z13 when portability and an integrated high-refresh touchscreen matter more than mini-PC expandability. In every case, compare the complete SKU, not just “Ryzen AI Max+ 395.”
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
