Estimate weight storage at common precisions from tensor shapes. Compare FP32, FP16 and INT8 weight footprints before choosing a device.
✳One place. Your next move.Add a file. We’ll find the tools.
PDFJPG
01 / START HERE
Add files to estimate model memory
Drop your files here, or choose them below. Add up to 50 and work at your own pace.
SAFETENSORS · ONNX · GGUF
How to estimate model memory
Add SAFETENSORS, ONNX, GGUF files.
Run the tool and wait for processing to finish.
Inspect the estimate model memory result and download it, or reuse the output in another tool.
Supported scope and limits
Static parsing within a 32 MiB combined input budget. No inference, Python imports or pickle execution. Reported compatibility is structural, not a guarantee a framework will load a model. Weight-only estimate. Runtime memory also includes activations, KV cache, allocator overhead and framework state.