The StableLM project warehouse contains the StableLM series language models being developed by Stability AI. At present, 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 under development.
Stability AI is currently HuggingFace Spaces A demo version of the 7 billion parameter model, StableLM-Tuned-Alpha-7B, has been released on. Users can visit this website to try it out.
StableLM Alpha model is trained on the basis of the expanded version of the Pile dataset, which contains 1.5 trillion tokens. The expanded version dataset is about three times the size of the Pile.
Size |
StableLM-Base-Alpha |
StableLM-Tuned-Alpha |
Training Tokens |
Parameters |
Web Demo |
3B |
checkpoint |
checkpoint |
800B |
3,638,525,952 |
|
7B |
checkpoint |
checkpoint |
800B |
7,869,358,080 |
HuggingFace |
15B |
have in hand |
(pending) |
|
|
|
30B |
have in hand |
(pending) |
|
|
|
65B |
have in hand |
(pending) |
|
|
|
175B |
(planned) |
|
|
|
|