File	Quantization	Size
`Holo-3.1-4B-Coding-Repair38-F16.gguf`	F16 converted GGUF	8,424,393,088 bytes
`Holo-3.1-4B-Coding-Repair38-Q8_0.gguf`	Q8_0	4,482,402,688 bytes
`Holo-3.1-4B-Coding-Repair38-Q6_K.gguf`	Q6_K	3,464,055,168 bytes
`Holo-3.1-4B-Coding-Repair38-Q4_K_M.gguf`	Q4_K_M	2,708,803,968 bytes

Notes

Q4_K_M is the supported 4-bit K-quant produced for this release by the available llama.cpp quantizer. No Q4_K_L file is published in this repository.

The source merged model, LoRA adapter, and tokenizer/config assets are maintained separately from this GGUF repository. This repository contains only the GGUF runtime artifacts and this model card.

Evidence files

Run evidence for this release is stored in the repository under evidence/:

These files are compact local/Kaggle run artifacts used to document training, evaluation, merge, or quantization evidence for this model family.

josephmayo/Holo-3.1-4B-Coder-GGUF

Holo-3.1-4B-Coding-Repair38 GGUF

Files

Notes

Evidence files

josephmayo/
Holo-3.1-4B-Coder-GGUF