DeepSeek R1 32B vs Mac Mini Pro 24GB | Install & Test

Install and testing of the 32B Parameter distilled model of DeepSeek R1 on base model Mac Mini M4 Pro, which has 24GB of RAM. This shows the model running on the Mac Mini and the kind of usable performance you can expect from it.
Useful Links:
DeepSeek R1 32B Test: www.head4space...
Granting Local DeepSeek R1 Internet Access: • Grant DeepSeek R1 Inte...
Ollama: ollama.com/
DeepSeek: www.deepseek.com/
Gear I Use: www.head4space...
🖥️ Complete Guide: Running Deep Seek R1 (32B) on Mac Mini M4 Pro
Watch as we walk through the step-by-step process of downloading, installing, and running Deep Seek R1's 32 billion parameter model on a base model Mac Mini M4 Pro (24GB RAM, 512GB SSD).
In this walk through, you'll learn:
How to download and install Deep Seek R1
Running the 32B parameter model locally
Performance testing with poetry and song lyrics
Real-time reasoning capabilities demonstration
Key Highlights:
✓ Complete installation walkthrough
✓ Performance metrics and GPU monitoring
✓ Real-world prompt testing
✓ Deep Seek's unique reasoning visualization
Perfect for developers and AI enthusiasts interested in running powerful language models locally on Apple Silicon hardware.
Want to see Deep Seek R1 with internet access? Check out this video: • Grant DeepSeek R1 Inte...

[@brianburton7716]

another nice generic question: "Who are the villains in the Die Hard series of movies." Local models are a mix of facts and hallucinating depending on the size of the model, the stupidly large online web models usually get the answer exactly right.

[@3dims]

That’s a good shout. I’ll keep it in mind for future experiments.