This is a refined version of my Rosa line, utilizing the same datasets as Foundation, but with a more sane approach to the creation of the LoRAs. This should be an improvement in every way.
The base of this model is Rosa v1, which should provide a more coherent end product than v3 after some extensive testing. It is my hope that this model will serve well as a character driven personal assistant/companion for users with low end hardware.
Feel free to leave commentary in the Community tab. I am always looking for new ways to improve my work, but for now, I consider this the realization of the goals for this project.
- Downloads last month
- 10
Inference API (serverless) does not yet support model repos that contain custom code.