this post was submitted on 21 Oct 2025
2 points (75.0% liked)

AI - Artificial intelligence

151 readers
16 users here now

AI related news and articles.

Rules:

founded 5 months ago
MODERATORS
Vacant@programming.dev
2
Experimenting with local LLMs on macOS (blog.6nok.org)
submitted 1 week ago by codeinabox@programming.dev to c/Aii@programming.dev
4 comments fedilink hide all child comments
top 4 comments
sorted by: hot top controversial new old
[–] onlinepersona@programming.dev 1 points 1 week ago (1 children)

Wat I don't understand is how it's supported on Mac. Do they have NVIDIA cards? I thought everything was dominated by CUDA. Did the lock-in company implement CUDA?

[–] codeinabox@programming.dev 2 points 1 week ago (1 children)

Ollama uses the Metal API on Apple Silicon Macs for GPU acceleration.

[–] onlinepersona@programming.dev 1 points 1 week ago (1 children)

Interesting. Is that similar to Vulkan?

[–] codeinabox@programming.dev 3 points 1 week ago

Vulkan?

I'm not too familiar with either, but this article goes into more detail: A Comparison of Modern Graphics APIs