L
localai.computer
ModelsGPUsSystemsBuildsOpenClawMethodology

Resources

  • Methodology
  • Submit Benchmark
  • About

Browse

  • AI Models
  • GPUs
  • PC Builds
  • AI News

Guides

  • OpenClaw Guide
  • How-To Guides

Legal

  • Privacy
  • Terms
  • Contact

© 2026 localai.computer. Hardware recommendations for running AI models locally.

ℹ️We earn from qualifying purchases through affiliate links at no extra cost to you. This supports our free content and research.

  1. Home
  2. Systems
  3. Minisforum EliteMini UM890 Pro

Minisforum EliteMini UM890 Pro

MinisforumMini PCRadeon 890M (integrated) · 16GB VRAM

Complete specifications and purchasing guidance for this pre-configured system.

Quick answer

Minisforum EliteMini UM890 Pro is best for teams that want a pre-configured path to local AI without building from parts.

Price

$799

CPU

AMD Ryzen 9 8945HS

GPU

Radeon 890M (integrated)

Memory

96GB


What's Inside
Hardware included with this configuration.
CPU
AMD Ryzen 9 8945HS
GPU
Radeon 890M (integrated)
Memory
96GB
Storage
2TB (2,048GB SSD)
Power supply
65W PSU
Discrete GPU included
Yes
Specifications
Technical details for deployment planning.
FieldDetails
ManufacturerMinisforum
CategoryPre-built system
Form factorMini PC
Total VRAM / unified memory16GB
GPU cores (aggregate)768
Power / TDP65W
Noise levelSilent
Dimensions130 x 126 x 49 mm
Warranty1 year
Release dateFeb 15, 2024
Where to Buy
Pre-configured systems available from authorized retailers.
Official StoreRecommended
Amazon

Note: Affiliate links help support LocalAI Computer. Prices may vary.

System decision workflow

Check model requirementsValidate compatibilityCompare GPU optionsReview build plansOpen buying guides

Systems FAQ

Is Minisforum EliteMini UM890 Pro good for local AI workloads?

Use this page as a baseline for memory, GPU, and power. Then validate exact model and quantization fit in compatibility checks before buying.

How should I compare this system against other options?

Compare price, available memory, and power draw against other systems and GPU pages, then map that to your target model requirements.

What is the next step after reviewing system specs?

Open model requirements and compatibility routes to confirm whether your target models run at acceptable speed and memory headroom.