Free local LLM tools.
Everything you need to pick, size and budget a local LLM: all free, no sign-up, built on our catalog of open models.
Which LLM for my PC?Enter your GPU or Mac and RAM: see which open models run on it, at which quantization.Which LLM can my Mac run?Pick your Apple chip and memory: the models that fit, their quantization and speed.LLM VRAM calculatorParameters, quantization and context in, the memory a model needs out.Quantization recommenderEnter your memory: the catalog models that fit, each at its best quantization.Local LLM budget plannerEnter a budget: machines that fit it, what they run, prices checked every week.LLM cost calculatorAPI bill vs self-hosting: monthly cost, electricity, amortization and break-even.API cost predictorYour API bill today and in 12 or 24 months, and how fast a local setup pays back.Local LLM leaderboardEvery locally runnable model ranked with a published, data-derived method.Model comparisonTwo local models side by side: size, memory, license, benchmarks.GPU comparisonTwo graphics cards compared for local AI: memory, bandwidth, what they run.Top 3 widgetEmbed the best local models for a given GPU on your own site.Public APIThe model catalog as JSON, CC BY 4.0, no key needed.Chrome extensionThe catalog offline in your browser, with a two-model comparison.
Model data comes from the BestLLMfor catalog (parameters and memory per quantization for every model we track). Hardware prices in the budget planner are checked every week by our offer checker. How we rank models: methodology.