What the calculator evaluates
LLMGauge accepts a hardware profile, a catalog model, and a quantization candidate. It estimates weight memory from parameter count and bits per weight, adds named overhead assumptions, and compares the result with dedicated VRAM and system RAM.
The recommendation workflow repeats the same engine evaluation across the curated catalog. The UI does not contain a second formula or a hidden ranking model.
What the result categories mean
These labels describe memory planning, not speed, quality, driver compatibility, or a guarantee that a particular downloaded file will run.
- GPU-capable: the estimate fits the available dedicated GPU memory under the current policy.
- Partial offload: the full estimate does not fit dedicated VRAM, but the combined memory picture allows a conservative split between GPU and system memory.
- CPU-only: the model is planned for system memory without relying on a discrete GPU, including the current conservative treatment of integrated graphics.
- Unsupported: the candidate exceeds the available memory pools or has no valid quantization candidate that fits.
Why an estimate is not a guarantee
Real requirements vary with the exact model file, quantization conversion, context length, runtime version, backend, driver, operating system, display use, and other programs competing for memory.
The initial LLMGauge policy is intentionally transparent: raw weights are estimated from parameters and bits per weight, a named weight overhead and runtime overhead are added, and system RAM reserves are included. Those assumptions are useful for comparison but are not benchmark measurements.
Use the result to choose a sensible starting point, then verify the exact model and runtime on your own system. A result near the boundary deserves extra headroom.