How fast can a local LLM be on my hardware?
A free tool that compares speed, memory, electricity costs, and total cost of local LLMs, image, and video generation across Mac, NVIDIA, DGX, and AMD hardware.
A lot of equipment being compared
- Mac Mini M4 16GB Local LLM Speed and Recommended Model
- Mac Mini M4 24GB local LLM performance and appropriate model
- Mac Mini M4 32GB Local LLM 27B Running Performance
- Mac Mini M4 Pro 48GB Local LLM 27B·35B Performance
- Mac Studio M4 Max 128GB local LLM large model performance
- RTX 3090 24GB local LLM used cost-effectiveness and performance
- RTX 4090 24GB local LLM and image creation performance
- RTX 5090 32GB local LLM/image/video generation performance
- Ryzen AI Max+ 395 128GB local LLM cost-effectiveness
- NVIDIA DGX Spark 128GB Local LLM Selection Criteria
Run in your browser
If the browser supports WebGPU, you can download a small language model directly to the device and generate responses to your questions without relying on a server.