Rig reports. Nothing estimated.
Every other number on this site is a prediction from memory bandwidth, honestly labelled as one. These are not. Each report below is a physical machine we rented, on which each model was actually loaded and timed — decode and prefill, TTFT at the 95th percentile, the VRAM it really took, how it degrades under concurrent users, watts at the wall, and what a million tokens costs. The llama.cpp build and the driver version are recorded so the numbers can be reproduced, or argued with.
Renting the machine is the cheap part. What takes the time is knowing which axes are worth sweeping — context depth, the quantisation ladder, the offload curve, concurrency — and building a harness that does not quietly measure itself instead of the hardware. Reports are re-run when llama.cpp moves enough to change the answer, and each one carries the build it was measured on, so a stale number is visible rather than hidden.
NVIDIA GeForce RTX 4090
17 models measured →NVIDIA GeForce RTX 3090
17 models measured →These machines were rented and measured because we wanted the numbers. If you need the same for hardware we have not run — your models, your context lengths, your concurrency ceiling, your quantisations — that is work we take on commission.
You would get exactly what is on this page: every measurement, the raw JSON, the host it ran on, the build it was measured against — and the results that do not flatter anyone, because the alternative is a report nobody can cite. Tell us the machine and the workload and we will say whether it is worth measuring at all.
ASK ABOUT A REPORT →Every VRAM calculator on the internet runs the same roofline we do and tells you to go validate it yourself. We went and validated it. The measurements here are also what corrects the estimator: the context curve, the offload curve and the KV-cache formula on this site were all rewritten from these files after they disagreed with the machine. How the estimates work → What a commissioned report contains →