LLaMA and other large language models locally on iOS and MacOS.
Various inferences and sampling methods.
Various inferences and sampling methods.
Simplicity and speed of development. One code for all devices. Low entry threshold.
View on github
The core is a Swift library based on llama.cpp, ggml and other open source projects that allows you to perform various inferences. A class hierarchy has been developed that allows you to add your own inference.
View Core repo
You can download this models from huginfaces
Popular model from Meta
RWKV-4 "Raven" models finetuned on Alpaca, CodeAlpaca, Guanaco, GPT4All, ShareGPT and more. Even the 1.5B model is surprisingly good for its size.
This is a model, which are GPT-2 models intended to generate prompt texts for imaging AIs.
The Cerebras-GPT family is released to facilitate research into LLM scaling laws using open architectures and data sets and demonstrate the simplicity of and scalability of training LLMs on the Cerebras software and hardware stack.
StableLM-Tuned-Alpha is a suite of 3B and 7B parameter decoder-only language models built on top of the StableLM-Base-Alpha models and further fine-tuned on various chat and instruction-following datasets.
