Tommy van der Vorst

> As i don't want to spam this thread further, is there a way to reach you directly (e.g. discord) if i have further questions? And again sorry for beeing...

Hi @raphaelmenges, happy to help! The issue with not returning the correct attribute may be related to the ONNX file you are using to test. Therefore I would suggest to...

> I have added a simple test for slice that should reproduce the first example of the ONNX specification of slice. I am still a bit unsure about the data...

Hi @philpax, thanks for bringing this up here. I have given the idea of running LLM’s through wonnx some thought over the past few days and I think it would...

> That sounds fantastic! Glad to see you're as interested as I am 🙂 > > > A way to load tensors (initializers in ONNX parlance) from GGML format. This...

I took a quick look at the ops in bold and I think most will be rather easy to implement. Some ops may not even be needed: * All `cont`...

> Agreed - there are lots of LLaMA models out there, but it's best to go for something unburdened. I'd suggest something like the RedPajama models, which are based on...

> @pixelspark Here are some converted [repajama models](https://huggingface.co/LLukas22/redpajama-ggml) which should work with the latest `main` branch. (I havent created the readme yet). That link shows a 404 for me?

Thanks @LLukas22, will try this later. If it works I can start investigating the different ops it uses (as listed by @philpax) and check if we can implement those.

> Apologies for the confusion there, it's been a bit hectic. We now target GGJT v3/QNT2 exclusively, as of five minutes ago 😅 So I finally got `llm` working with...