MaziyarPanahi

/

yuna-ai-v4-full-GGUF

like 0

Hi @MaziyarPanahi
Would you mind chiming in here?
@yukiarimo Thank you for reporting, if you are not able to come to a resolution here, could you please forward us a DMCA takedown notice so we can help you move this along?

Hi @yukiarimo & @no-mad

• The GGUF file in MaziyarPanahi/yuna-ai-v4-full-GGUF was produced by running the standard llama.cpp quantize utility on the original model (before it was gated of course).
• That upstream commit already carried the Creative Commons BY-NC-ND 4.0 license. Under # 2 (a)(4) the license explicitly states that technical modifications necessary for exercise of the licensed rights “never produces Adapted Material.” Quantisation is only a numeric precision change, analogous to compression, and is necessary for use on CPU devices.
• I have redistributed the model unchanged in content, for non-commercial purposes, with full attribution and a link back to the original repo, satisfying BY-NC-ND requirements.

Therefore, the GGUF file remains licensed under CC BY-NC-ND 4.0 and its redistribution is permitted. If you still believe this interpretation is incorrect, please feel free to submit a formal DMCA takedown notice to Hugging Face. Once a notice is received, the platform is legally obligated to remove the content. At that point, I would be entitled to file a DMCA counter‑notice, fully outlining how quantization is a permitted technical change, that my distribution was non‑commercial and properly attributed, and that no derivative material was created.
I’m also happy to assist with reviewing either notice to ensure clarity.

Either remove the model completely, because I have official quantization or include official README.md and LICENSE.md files with 0% modification

Hi @MaziyarPanahi
Would you mind chiming in here?
@yukiarimo Thank you for reporting, if you are not able to come to a resolution here, could you please forward us a DMCA takedown notice so we can help you move this along?

How to submit it? Can you please give me a step-by-step guide? Can't find it on their website!

Either remove the model completely, because I have official quantization or include official README.md and LICENSE.md files with 0% modification

This is doable of course. I will add them both tomorrow.

@yukiarimo since the repo is gated, i don't have access to those files. would you mind sharing your README and LICENSE files here so I can include them in this repo?

Hi @yukiarimo , can I ask what is the base model that the yukiarimo/yuna-ai-v4-full model is finetuned from? From the filesize of the GGUF quantizations, it appears that it might be derived from Llama 3, and if so, you should abide by the terms of the Llama 3 Community License Agreement or the Llama 3.1 Community License Agreement.

However, I do not see any prominently displayed attribution in the model card, only a reference that the model uses the "LLaMA 3.1 8B" architecture. Also, the model name does not include "Llama" anywhere within it, although it is required according to the license terms.

And I am not sure if you are allowed to give your derivative model another license, instead of inheriting the original license of the base model. The original Llama 3 license terms state that you need to "include a copy of the agreement", but I am unsure if it means just putting the license file in the repo, or the finetuned model also has to use the same license.