GGML missing FF tensor assertion PoC

This repository contains a minimal legacy .ggml file that makes the public llama.cpp converter convert_llama_ggml_to_gguf.py abort with:

AssertionError: Missing layer 0 FF tensor

Files

malicious_missing_ff_tensor.ggml Tiny malformed GGML file with valid header and vocab but no tensors.
build_poc.py Deterministically rebuilds the file.
reproduce.py Runs the public converter on the file and captures the assertion.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support