Beelink SER9 Pro AMD Ryzen™ 7 H 255 (OpenClaw Pre-Installed)
Beelink SER9 Pro AMD Ryzen™ 7 H 255 (OpenClaw Pre-Installed) Price range: $759.00 through $1,029.00
Back to products
Beelink SER8 AMD Ryzen™ 7 8845HS
Beelink SER8 AMD Ryzen™ 7 8845HS Price range: $799.00 through $1,239.00

Beelink SER9 PRO AMD Pro Ryzen™ AI 9 HX 370 (OpenClaw & Local LLM Pre-installed )

(5 customer reviews)

Price range: $1,169.00 through $1,389.00

12 Zen5 CPU cores, combined with a new Radeon™ 890M iGPU, a new generation XDNA 2 NPU and 80 AI TOPS, the Ryzen™ AI 9 HX370 processor’s fully upgraded architecture significantly boosts the SER9’s overall performance, delivering a faster, smarter, and more secure Al PC experience!.

19 Items sold in last 7 days
22 People watching this product now!
Estimated delivery: 7-14 days
SKU: BL-SER9PRO-HX370-OC Category: Tags: , , , ,
Description

The Beelink SER9 PRO pairs AMD’s Ryzen AI 9 HX 370 — 12 Zen 5 cores, a Radeon 890M iGPU, and an XDNA 2 NPU good for 80 AI TOPS — with OpenClaw and local LLM tooling pre-installed, all inside the same compact 135 x 135 x 44.7mm chassis as Beelink’s other SER-series machines. It’s built for developers and AI enthusiasts who want genuine on-device model inference in a machine small enough to disappear on a desk, not a full tower workstation.

Key Features

  • AMD Ryzen™ AI 9 HX 370 — 12 Cores, 24 Threads — Zen 5 cores boost up to 5.1GHz with 36MB of combined L2+L3 cache, giving the SER9 PRO strong single- and multi-thread performance for compiling, running local models, and heavy multitasking simultaneously.
  • AMD Radeon™ 890M — RDNA 3.5, 16 Compute Units — Running at up to 2900MHz, the 890M is widely regarded as the fastest integrated GPU class available in a mini PC, capable of handling GPU-accelerated inference workloads alongside genuine light gaming.
  • 80 AI TOPS via XDNA 2 NPU — A dedicated NPU delivers 80 TOPS of AI acceleration, giving OpenClaw and local inference tooling like Ollama or llama.cpp real on-device throughput for quantized models, rather than depending entirely on a cloud connection.
  • 32GB LPDDR5X 7500MHz Memory — High-bandwidth soldered LPDDR5X memory keeps model loading and inference responsive, sized for running mid-size quantized open-weight models locally alongside your normal workload.
  • Dual PCIe 4.0 NVMe Storage, Up to 8TB — Two M.2 2280 PCIe 4.0 x4 slots support up to 8TB combined, with the retail unit including a 1TB or 2TB Crucial SSD, leaving room to store multiple model checkpoints alongside your OS and projects.
  • Triple 4K Display Output — DisplayPort 1.4 (4K@240Hz), HDMI 2.1 (4K@240Hz), and USB4 (40Gbps/TB3/PD/DP1.4) together support up to triple 4K displays for serious multi-monitor productivity setups.
  • 32dB Near-Silent Cooling — Efficient bottom-intake airflow keeps the chip stable at 65W TDP with very low fan noise, so it stays quiet even while running sustained AI inference workloads.
  • AI-Enhanced Voice Pickup and Dual Speakers — A built-in microphone processed through an AI chip delivers 360° barrier-free voice recognition within 5 meters, paired with dual speakers for calls and media without external audio gear.

Why Choose the SER9 PRO

The HX 370’s 80 TOPS NPU is the real differentiator here — it’s enough dedicated AI silicon to make local inference genuinely practical, not just a checkbox spec. For developers building agentic coding workflows with OpenClaw, that means the option to run smaller quantized models entirely on-device: no API latency, no per-token cost, no sending code or prompts to a third party. Paired with 32GB of fast LPDDR5X memory, it’s a real local AI development box in a footprint that fits on a shelf.

It’s equally capable as a general-purpose workstation. The 890M iGPU is fast enough for creative and light gaming workloads, triple 4K display support covers serious multi-monitor setups, and the 12-core HX 370 handles everyday multitasking without hesitation — this isn’t a machine you compromise on for AI capability.

If you need to run the largest local models available — 70B-class and above — our GTR9 Pro’s 128GB unified memory platform is the better fit. The SER9 PRO’s sweet spot is mid-size quantized models and agentic coding workloads in a genuinely portable, quiet, affordable package, without needing the higher-tier AI Max+ hardware. Every SER9 PRO ships from our UK warehouse with 1-3 business day processing, free shipping over $199, and ComputaHardware support if you need help getting your local LLM setup dialed in.

Specifications

  • CPU: AMD Ryzen™ AI 9 HX 370, 12C/24T Zen 5, up to 5.1GHz boost, 36MB cache
  • GPU: AMD Radeon™ 890M, 16 CUs, up to 2900MHz, RDNA 3.5
  • AI: XDNA 2 NPU, 80 AI TOPS
  • Memory: 32GB LPDDR5X 7500MHz (soldered)
  • Storage: Dual M.2 2280 PCIe 4.0 x4 slots, up to 8TB combined; 1TB or 2TB Crucial SSD included
  • Display outputs: DisplayPort 1.4 (4K@240Hz), HDMI 2.1 (4K@240Hz), USB4 (40Gbps/TB3/PD/DP1.4) — triple 4K support
  • Networking: Wi-Fi, Bluetooth
  • Cooling: ~32dB near-silent operation, 65W TDP
  • Audio: AI-enhanced 360° 5-meter microphone pickup, dual speakers
  • Design: Dust-proof bottom mesh filter
  • Software: OpenClaw and local LLM tooling pre-installed; Windows 11 Pro or Ubuntu
Additional information
RAM+SSD32GB LPDDR5x memory (soldered) + 1TB SSD, 32GB LPDDR5x memory (soldered) + 2TB SSD
Operating SystemUbuntu, Ubuntu + Windows
Power AdapterAU, EU, JP, UK, US
Reviews (5)

5 reviews for Beelink SER9 PRO AMD Pro Ryzen™ AI 9 HX 370 (OpenClaw & Local LLM Pre-installed )

  1. Sinead C. –

    Really solid mini PC. The built-in dual speakers are actually decent for casual use, didn’t expect much but they’re clear enough for calls and background music. Local LLM setup was mostly done for me out the box.

  2. Petra L. –

    Good value for the chip on offer. I mainly use it for general productivity with some local AI experimentation on the side. One small gripe, the power LED is quite bright and visible from across the room at night, wish there was a way to dim it.

  3. Milo R. –

    Plenty of AI performance for the local models I run day to day through OpenClaw, mostly smaller coding assistant models. Runs at a low TDP and stays impressively quiet, I barely notice it under my desk. Great purchase.

  4. Thiago B. –

    Been using this as a dedicated OpenClaw agent host for a few weeks now, handling small automation and coding tasks reliably. Thermal performance is excellent, barely warm to the touch even after hours of continuous use.

  5. Roland K. –

    Machine itself performs well but mine arrived with a scuff on the top panel, clearly minor cosmetic damage from shipping rather than a defect. Contacted support and they were helpful, offered a partial refund. Performance wise no complaints.

Add a review

Your email address will not be published. Required fields are marked *

Shipping and Delivery

Fast, Tracked Worldwide Shipping

In-stock orders are processed within 1–3 business days and shipped with full tracking. Delivery costs and timeframes depend on your destination:

Delivery Estimates & Costs

  • United Kingdom — 2–4 business days, $9.95 (free over $199)
  • Europe — 5–10 business days, $19.95 (free over $199)
  • United States & Canada — 6–12 business days, $24.95 (free over $199)
  • Rest of world — 10–20 business days, $49.95 (free over $499)

Free shipping is applied automatically at checkout once your order passes the threshold for your destination — no code needed. Every order ships with a tracking number emailed to you as soon as it leaves our warehouse.

International orders may be subject to import duties or customs fees set by your local authorities; these are not included in the checkout price and are the buyer’s responsibility.

Read our full Shipping Policy for details on processing, customs, damaged or lost parcels, and delivery addresses.