Read metadata and tensor shapes from a GGUF.
How to use GGUF Inspector
Add a GGUF file. Drop a .gguf file onto the uploader or pick it from your device. Only one file is accepted at a time.
Wait for the header to be read. The tool reads the start of the file and shows Reading header while it parses.
Check the GGUF header summary. Read the format Version and Tensor count at the top.
Browse the metadata. Scroll the Metadata table of key-value pairs. Long values are truncated in the cell and shown in full on hover.
Browse the tensors. Scroll the Tensors table to see each tensor's name, shape and ggml type.
Limits and privacy
- Only the beginning of the file is read, never the tensor data. It starts with a 4 MB read and doubles the window up to a 64 MB ceiling if the header is larger.
- The upload limit is 50 GB per file, and only one file at a time.
- The file must have a .gguf or .safetensors extension. ONNX, .bin and .pt files are not supported by this inspector.
- The parser rejects headers that report more than 1,000,000 tensors or metadata entries, metadata arrays longer than 100,000 items, or tensor entries with more than 8 dimensions.
- Metadata arrays longer than 12 items are shown as the first 12 items plus the total count.
- Tensor types are looked up in a built-in ggml type list. A type outside that list is shown as unknown with its numeric value.
- Processing happens in your browser. The file is not uploaded.
Frequently asked questions
How do I see what quantization a GGUF file uses?
Open the file and look at the Type column in the Tensors table, which lists each tensor's ggml type such as Q4_K or F16. Metadata keys may also include a file type entry, depending on the file.
Does it load my whole model into memory?
No. It reads only the leading bytes of the file where the header lives, so large multi-gigabyte files open quickly.
Is my model uploaded anywhere?
No. The header is parsed locally in your browser.
Why does it say my file is not supported?
The inspector decides the format from the file extension and only accepts .gguf and .safetensors. If the extension is right but the file is not a real GGUF, it reports that the magic number does not match.