Find the Perfect Hardware
for Your LLM

Match any Hugging Face model to optimal hardware. Compare specs, benchmarks, and prices.

Get Recommendations

What Are You Building?

Choose a use case to see tailored hardware recommendations.

💻 Coding 🧠 Reasoning 🎨 Image Generation 🎬 Video Generation 🔊 Speech / TTS 🤖 General Purpose

Popular Models

Trending models on Hugging Face right now.

View All Models →
Meta

Llama 3.3 70B Instruct

70B params | 5.0M downloads
🤖 General Purpose
Stability AI

Stable Diffusion XL Base 1.0

3.5B params | 5.0M downloads
🎨 Image Generation
Mistral AI

Mistral 7B Instruct v0.3

7B params | 4.0M downloads
🤖 General Purpose
Meta

Llama 3.2 8B Instruct

8B params | 3.5M downloads
🤖 General Purpose
DeepSeek

DeepSeek R1

671B params | 3.0M downloads
🧠 Reasoning
Qwen

Qwen2.5 72B Instruct

72B params | 2.5M downloads
🤖 General Purpose 🧠 Reasoning
Meta

Llama 3.2 3B Instruct

3B params | 2.0M downloads
🤖 General Purpose
Black Forest Labs

FLUX.1 Schnell

12B params | 2.0M downloads
🎨 Image Generation

How It Works

Three simple steps to find your ideal LLM hardware.

1

Select Your Model

Choose from hundreds of Hugging Face models or filter by use case, parameter count, and provider.

2

See Requirements

We calculate the memory, bandwidth, and compute needed at your chosen quantization level.

3

Buy Hardware

Get ranked hardware recommendations with direct purchase links from top retailers.

Featured Hardware

Top picks for running local LLMs in 2025.

View All Hardware →

Mac Mini M4 Pro (48GB)

$1,599

Memory

48GB unified

Bandwidth

273 GB/s

CPU Cores

14

GPU Cores

20

Performance 6/10

NVIDIA GeForce RTX 4090 (24GB)

$1,599

Memory

24GB gddr6x

Bandwidth

1008 GB/s

CPU Cores

N/A

GPU Cores

16384

Performance 8/10

NVIDIA GeForce RTX 5090 (32GB)

$1,999

Memory

32GB gddr7

Bandwidth

1792 GB/s

CPU Cores

N/A

GPU Cores

21760

Performance 10/10