Running Local AI on AMD

Short summary

Sam Witteveen walks through running local LLMs and image/video models on AMD GPUs using LM Studio, ROCm, Unsloth, and ComfyUI. Covers hardware requirements, token-cost advantages vs. cloud APIs, and practical setup on Windows and Linux. Includes GitHub tutorials and AMD Developer Program resources.

• •Tutorial on running LLMs locally on AMD hardware using LM Studio, ROCm, Unsloth
• •Compares AMD GPU vs cloud API costs for local inference workflows
• •Practical demos with ComfyUI and tools; better performance on Linux

Generated with AI, which can make mistakes.