"Local models, real GPUs."
Discover how different AI models perform on real hardware. Same prompts, same rig, one variable changed per test.
Every test uses identical prompts to ensure fair comparison between models.
All tests run on the same hardware configuration for consistent results.
Only one variable changes per test: the AI model being evaluated.
The model is reloaded before every single run to ensure fresh starts.
Every prompt and result file is published for complete reproducibility.
NVIDIA RTX 3090
LM Studio
A YouTube series dedicated to testing AI models on real, local hardware. We believe in transparency, reproducibility, and showing you exactly what different models can do when running on your own GPU.