Any model fits your GPU.Too big for your GPU? It shrinks to fit, recovers from crashes, and remembers what worked.
Free during the beta — no account, no sign-up.
VRAMPilot is a free local tool that runs GGUF models with llama.cpp and recovers from out-of-memory at runtime instead of crashing. You point it at a .gguf file; it reads your GPU, picks a configuration that fits, launches a llama-server, and if the model still runs out of memory — at boot or in the middle of a generation — it backs off, retries until it serves, and reports what it traded away to make the model fit.
It is a UX/automation layer on top of llama.cpp, which does the actual inference: if llama.cpp cannot run your model, neither can VRAMPilot. Read the docs →
Too big for your VRAM? It shrinks to fit — quant, context, offload.
An out-of-memory is caught and recovered. The server stays up.
Your own hardware — never a rented GPU.
Fits models onto the GPU you already have instead of renting more — less hardware, less waste.
Fits a model over your VRAM budget.
by designCatches out-of-memory and keeps serving.
measuredNVIDIA, AMD, Apple — or plain CPU.
verifiedEvery number is measured and public. Read the raw runs and reproduce them yourself.
vrampilot is built by ZMLabs — a deep-engineering studio in Sète, South of France, making powerful software accessible.
Point it at a model — it adapts and keeps serving.