2 ms·You can do LLM inference without CUDA just fine. Download Ollama and see for yourselfby adam_arthur 3y agoYou can do LLM inference without CUDA just fine. Download Ollama and see for yourself