UPSTREAM PR #21095: convert: Add compressed-tensors NVFP4 conversion

[loci-dev]

Note

Source pull request: ggml-org/llama.cpp#21095

This update expands the convert_hf_to_gguf script to support converting Huggingface NVFP4 models quantized with compressed-tensors. Previously, only ModelOpt quantized models were compatible and an error was raised.

It finds the values and names used by compressed-tensors (eg, weight_global_scale instead of weight_scale_2 for the tensor scale) and renames them to the ModelOpt equivalents so that the rest of the conversion remains identical. This keeps the update small. The weights themselves do not need any adaptation; the only other difference is that the scales become reciprocal values.

[loci-review]

No meaningful performance changes were detected across 123908 analyzed functions in the following binaries: build.bin.libllama.so, build.bin.llama-tts, build.bin.llama-cvector-generator, build.bin.llama-bench, build.bin.libmtmd.so, build.bin.libggml.so, build.bin.libggml-cpu.so, build.bin.libggml-base.so, build.bin.llama-gemma3-cli, build.bin.llama-gguf-split, build.bin.llama-llava-cli, build.bin.llama-minicpmv-cli, build.bin.llama-quantize, build.bin.llama-qwen2vl-cli, build.bin.llama-tokenize.

🔎 Full breakdown: Loci Inspector
💬 Questions? Tag @loci-dev