Don’t have enough VRAM for some local models yet?

Short Course Q&A AI Coding Workflows: From Cloud to Local
1

If you don’t want to run locally for some reason or don’t have the vram, etc, The smaller models can be executed via api at hugging face inference providers: Inference Providers · Hugging Face

2

Cool, thanks for sharing!

3

There is still an option to run them with RAM. GitHub - AlexsJones/llmfit: Hundreds of models & providers. One command to find what runs on your hardware. · GitHub can help detect what you can run on your system.