MyLocalRig articles

Make the next local AI decision.

Choose your situation below. You will either get a hardware-specific result, a model list, or a focused guide with a concrete decision at the end.

Already know your hardware?

Enter the exact GPU or Apple chip, RAM, and free storage to see suitable models.

Start with your situation

What are you trying to do?

Find models that fit its accelerator memory, RAM, storage, and operating system.

Compare popular models by purpose, release date, size, context, memory, and license.

Work backward from the model and workload before choosing parts.

Choose a runtime, load a manageable model, and verify where it is running.

Focused explanations

These articles explain decisions the model table and rig checker cannot make for you.

Hardware planning7 min

Turn a workload into a model class, memory target, platform choice, and balanced parts list.

Memory planning5 min

Estimate weight memory, account for context and overhead, and decide whether CPU offload is acceptable.

First run6 min

Choose among LM Studio, Ollama, and llama.cpp, then confirm the model uses the expected hardware.

Evidence standard