Build My Rig
Budget

Starter AI Desktop

Run your first local AI models for under $600

$543

7 components · 12 GB VRAM · 6 compatible models

VRAM

12 GB

TDP

250W

Noise

~35dB

Models

6

Tier

Budget

The Heart of This Build

NVIDIA GeForce RTX 3060 12GB

$269

12GB VRAM at the absolute lowest price. Runs 3B-8B models at Q4-Q5 comfortably. The used market makes this even cheaper.

Buy New
AmazonBest BuyNeweggB&H Photo
Buy used: save ~$94
eBayMarketplacer/hardwareswap
Parts

7 components, $543 total

Estimated prices. Actual retail may vary by region. Some links may earn a small affiliate commission.

The Rest of the Build

cpu

Intel Core i3-12100F

4-core budget CPU. Inference is GPU-bound, so CPU matters less. Enough for system tasks while GPU handles AI.

$79
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
motherboard

MSI PRO B660M-A WiFi

Basic B660 mATX board with WiFi. PCIe 4.0 x16 for the GPU. No frills needed.

$89
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
ram

16GB DDR4-3200 (2x8GB)

16GB is tight but workable for entry-level. Models load to GPU VRAM, not system RAM for inference.

$32
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
storage

500GB WD Blue SN580 NVMe

500GB holds 10-15 quantized models. Upgrade later if needed.

$35
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
psu

EVGA 500 BR 500W 80+ Bronze

500W is sufficient for RTX 3060 (170W TDP). Budget-friendly with adequate efficiency.

$39
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
case

Thermaltake Versa H18

Cheap, functional micro-ATX case with adequate airflow. Gets the job done.

$39
Where to buy
Direct
AmazonBest BuyNeweggB&H Photo
Runs
Compatibility

What This Build Can Run

6 AI models benchmarked on this exact hardware configuration.

Fastest Model
120tok/s

Fast

3Fast (40+ tok/s)
2Good (25–39 tok/s)
1Slow (<12 tok/s)
Llama 3.2 3B Instruct
Q8_0
80
Phi-4 Mini
Q8_0
75
Llama 3.1 8B Instruct
Q4_K_M
30
Mistral 7B Instruct v0.3
Q4_K_M
30
Whisper Large V3
FP16
0
Next Step

Upgrade Path

Add 16GB more RAM ($32) for system headroom. Swap GPU to RTX 4060 Ti 16GB ($449) for a massive VRAM upgrade. The platform supports it without other changes.

Built For

Target Use Cases

ChatCoding

Want to tweak this build?

Open it in the configurator to swap components, check compatibility, and see what models you can run.

Customize This Build

Prices are estimates and may vary by retailer and region.