Beelink SEi13 AI and SEi14 AI Mini PCs Launch with 160 TOPS NPU

Beelink launches SEi13 AI and SEi14 AI mini PCs featuring independent NPUs with up to 160 TOPS compute power and 24GB dedicated AI memory.

Beelink SEi13 AI and SEi14 AI Mini PCs Launch with 160 TOPS NPU

Beelink introduced two new mini PCs designed specifically for local artificial intelligence workloads. The SEi13 AI and SEi14 AI models bring dedicated neural processing power to compact desktops. This shift matters because it allows users to run AI inference tasks locally without relying on cloud services. Buyers who need private, on-device AI capabilities now have dedicated hardware options that bypass the need for expensive GPU upgrades.

Beelink SEi13 AI and SEi14 AI mini PCs
Beelink SEi13 AI and SEi14 AI mini PCs

Dual-compute architecture targets local AI inference tasks

The company released the SEi13 AI and SEi14 AI on July 29, 2024. Both systems utilize an x86 CPU architecture paired with an independent NPU. This dual-compute setup separates standard processing tasks from AI-specific calculations. The design aims to optimize performance for reasoning-heavy applications by dedicating resources to the neural engine.

Spec comparison

Spec SEi13 AI SEi14 AI
Processor Intel Core i7-13620H Intel Core Ultra 9 185H
NPU Compute Up to 160 TOPS Up to 160 TOPS
AI Memory 24GB LPDDR5 24GB LPDDR5
Standard RAM 32GB 32GB
Standard Storage 1TB 1TB

The SEi13 AI features an Core i7-13620H processor. The higher-end SEi14 AI uses an Intel Core Ultra 9 185H chip. Both models offer up to 160 TOPS of compute power from the independent NPU. They also include 24GB of LPDDR5 memory dedicated to AI reasoning tasks.

Standard configurations for both units include 32GB of RAM and 1TB of storage. The SEi13 AI is priced at 11,895 CNY (around $1,756). The SEi14 AI costs 13,988 CNY (around $2,065). These prices apply to the Chinese market where the devices launched on the announcement date.

Beelink positions these machines as tools for local AI inference. The combination of x86 processing and independent NPU compute defines their technical approach. The 24GB LPDDR5 allocation ensures sufficient memory bandwidth for AI models. The 32GB system RAM supports general operating system and application needs.

Discussion

0 comments

Log in to join the thread with a thoughtful take, question, or correction.

Add to the discussion