fakezeta

Hi all, I'm currently on vacation with my family this week and as such I've limited capabilities to support (I'm only with mobile phone). From the last log I saw...

@mudler there could be something in the non-streaming generation since streaming is working. I'll look at it when I get home in the weekend. @gericho Ollama is using the same...

Back at home and working on this thanks to @gericho availability on direct chat. Happy to see that performance are consistent with my testing HW and it confirms my idea...

Thank you for your kind words, which truly warm my heart, especially considering the late-night hours I’ve spent on the transformer backend. Regarding the chat template issues, the pull request...

@nickp27 the PR is in his journey to be merged: I just opened it yesterday 😄 I tested this model for you with my local build and I think there...

PR merged, it's in the latest docker build `local-ai:master-sycl-f16-ffmpeg` On my [HF](https://huggingface.co/fakezeta) you can find some other models like WizardLM2 and also Llama3. Beware that I couldn't get Llama3 to...

Thank you @airMeng, I read the README but it's only giving warning on performance not functionality. Tested with OpenVINO, the same oneAPI version and the same models I had not...

Sorry for the late reply. Tested with a more recent build and I can confirm that it's working fine. Thank you.

Hi @adamcohenhillel, thank you for your interest in LocalAI. The discussion is on llama.cpp repo, not Ollama and LocalAI already uses llama.cpp as one of the backends so as soon...

Hi @richiejp thanks again for your time. This seems an upstream (openvino) error. The kernel log seems to be related to accessing a memory region not allowed. Like a null...