TinyKAI-0.7B is one of the world's SMALLEST "large" language models EVER! We acheived such a feat by essentially compressing GPT2 and squeezing it to fit into one sub-GB pytorch.bin file!

Due to being one of the tiniest "large" language models on huggingface, many corners had to be cut. TinyKAI-0.7B has a max output of only 150 characters per computation (without repeating itself). Therefore, TinyKAI is only intended for research purposes.

TinyKAI-1B is governed by the apache 2.0 liscense, and therefore means that whatever the license deems unacceptable shall not be allowed. We specificaly ban the use of ANY AND ALL KAI MODELS for hate speech towards a paricular thing, person, our particular group due to legal and ethical issues.

2K characters max output length, English only, and extremely tiny (1GB).

TinyKAI-0.7B is a very small model, and is good for AI research especially.