Local LLM Fit Calculator

Pick your GPU, system RAM, and context length — we compute weights, KV cache, and overhead to show exactly which models and quants fit and how fast they'll run.