Anthony
Anthony
I was able to resolve this by manually setting the chat template: ```python llava_1_5_chat_template = '''{% for message in messages %} {% if message.role == 'system' %} {{ message.content.strip() }}...
@okaneyo, you can test my attempt to implement Phi 3.5 MoE in https://github.com/ml-explore/mlx-swift-examples/pull/116 and tell me if it works. My MacBook unfortunately doesn't have enough RAM for me to test...
The main motivation would be discoverability, especially since models are now commonly ported with automated tools. This came to mind as I was cleaning up the mlx-audio repo, which contained...
Sorry about that. I ran the formatting again.
I don't have an SVG, but you can find a PNG [here](https://localchat.co/favicon.png). Would that work?
I'd love to get this merged. Is there anything else I need to do on my end?
It's been over two months since I submitted this PR. I'm an active contributor to several MLX and Hugging Face repositories in addition to creating one of the first iOS...
@Vaibhavs10, it's been 4.5 months since I submitted this PR. I would really love to get this merged so that users of Local Chat can start using MLX models directly...
Are you testing this with https://github.com/ml-explore/mlx-swift-examples/pull/338?
Before submitting my PR for review, I tested it and noticed a performance difference with the quantized KV cache, so I believe it was working for me. I don't have...