Local LLM Finder

Local model compatibility

Which local LLM can your computer run?

Choose your hardware and workload to compare local models by estimated memory use, context length and task fit. Speed estimates are available for supported GPUs.

Basic

Hardware

Computer memory (RAM)

Main use

Pick one

This gets the most weight.

Priority

Balanced
FasterStronger answers
Tune results8,000 tokens
Also useful for

Secondary uses influence ranking; main use stays primary.

How much text the estimate should account for, including prompt, pasted text and reply.

Model filtersAutomatic model choices

Automatic picks the best-fitting option. Fixed choices can exclude models.

Estimated quantized size before runtime/context overhead.

Leave open for the best overall match.

Replaces the memory budget; speed estimate is disabled.

Matches

Example: 16 GB RAM.

Catalogue Sep 2026

Choose hardware and a main use.

Read model requirements and guides

Results are planning estimates, not benchmarks. Check the model publisher, licence and runtime before relying on a result or buying hardware. How estimates work