RUNLOCALAIv38
->Will it run?Best GPUCompareTroubleshootStartLearnPulseModelsHardwareToolsBench
Run check
RUNLOCALAI

Independently operated catalog for local-AI hardware and software. Hand-written verdicts. Source-cited claims. Reproducible commands when we have them.

OP·Fredoline Eruo
DIR
  • Models
  • Hardware
  • Tools
  • Benchmarks
TOOLS
  • Will it run?
  • Compare hardware
  • Cost vs cloud
  • Choose my GPU
  • Prompting kits
  • Quick answers
REF
  • All buyer guides
  • Learn local AI
  • Methodology
  • Glossary
  • Errors KB
  • Trust
EDITOR
  • About
  • Author
  • How we make money
  • Editorial policy
  • Contact
LEGAL
  • Privacy
  • Terms
  • Sitemap
MAIL · MONTHLY DIGEST
Get monthly local AI changes
Monthly recap. No spam.
DISCLOSURE

Some links on this site are affiliate links (Amazon Associates and other first-class retailers). When you buy through them, we earn a small commission at no extra cost to you. Affiliate links do not influence our verdicts — there are cards we rate highly that we don't have affiliate relationships with, and cards that sell well that we refuse to recommend. Read more →

© 2026 runlocalai.coIndependently operated
RUNLOCALAI · v38
  1. >
  2. Home
  3. /Learn
  4. /Courses
  5. /Hardware Planning for Local AI
  6. /Ch. 11
Hardware Planning for Local AI

11. External GPU Enclosures

Chapter 11 of 20 · 15 min
KEY INSIGHT

eGPU enclosures provide upgrade paths for laptops at 15-25% performance cost versus internal GPUs—acceptable for development, marginal for production use. ```bash # Verify Thunderbolt connection on macOS system_profiler SPThunderboltDataType # Check eGPU enumeration on Linux ls -la /sys/bus/pci/devices/ # Identify the eGPU by vendor ID (NVIDIA = 10de) lspci | grep -i vga ```

External GPU (eGPU) enclosures provide upgrade paths for systems without internal GPU capability. This includes laptops, Mini-ITX builds, and compact form factors.

eGPU Architecture

eGPU setups consist of:

  1. Enclosure with desktop GPU
  2. Power supply (typically 500-700W built-in)
  3. Interface (Thunderbolt 3/4 or USB4)

Bandwidth constraints:

Interface Bandwidth eGPU Efficiency
Thunderbolt 3 40 Gbps 75-85%
Thunderbolt 4 40 Gbps 75-85%
USB4 20-40 Gbps 60-70%

Compatible Enclosures

Enclosure Max GPU Length Power Price
Razer Core X 12.4" 500W $300
Sonnet eGFX Breakaway 550 12.3" 550W $350
Akitio Node Pro 11.5" 600W $400
ASUS XG Station Pro 13" 500W $250

GPU length matters—RTX 4090 at 12.3" fits most enclosures, but longer cards may not.

Real Performance Numbers

Testing Llama 3 8B inference via eGPU:

Setup Tokens/sec Notes
Internal RTX 3060 22 Baseline
eGPU RTX 3060 (TB3) 19 14% slower
eGPU RTX 4070 (TB4) 24 Acceptable loss

The bandwidth overhead primarily affects batch processing. Single-query inference shows minimal degradation.

Setup Process

# Linux eGPU setup
sudo apt install nvidia-driver-535
echo 'auto eGPU' | sudo tee /sys/bus/pci/slots/1/power

# Verify GPU detection
nvidia-smi

# Common issue: Hot-plug failures
# Solution: Cold boot with eGPU connected
# 1. Shut down
# 2. Connect eGPU
# 3. Power on enclosure
# 4. Boot system

Failure Modes

  1. Hot-plug instability: Always connect eGPU before boot
  2. Thermal throttling: Enclosure fans may be insufficient
  3. Audio interference: Some enclosures introduce buzzing
  4. Driver conflicts: Multiple GPUs require careful routing
EXERCISE

Research three eGPU enclosures and compare them across GPU length support, power delivery, and Thunderbolt version. Identify the best option for your current system.

← Chapter 10
Motherboard and PSU
Chapter 12 →
Cloud GPU Fallback