The StableLM project repository contains Stability AI’s ongoing development of the StableLM series of language models, and currently Stability AI has released the initial StableLM-alpha model set with 3 billion and 7 billion parameters. Models with 15 billion and 30 billion parameters are in development.
Stability AI is currently at Hugging Face Spaces A demo version of the 7 billion parameter model, StableLM-Tuned-Alpha-7B, was released on , and users can visit the website to try it out.
The StableLM-Alpha model is trained on the expanded version of the Pile dataset, which contains 1.5 trillion tokens, and the size of the expanded version of the dataset is about 3 times that of the Pile.
| size | StableLM-Base-Alpha | StableLM-Tuned-Alpha | Training Tokens | Parameters | Web Demos |
|---|---|---|---|---|---|
| 3B | checkpoint | checkpoint | 800B | 3,638,525,952 | |
| 7B | checkpoint | checkpoint | 800B | 7,869,358,080 | Hugging Face |
| 15B | in progress | (pending) | |||
| 30B | in progress | (pending) | |||
| 65B | in progress | (pending) | |||
| 175B | (planned) |
#StableLM #Homepage #Documentation #Downloads #Language #Model #Developed #Stability #News Fast Delivery