Instructions to use fulcrumresearch/echo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use fulcrumresearch/echo with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
|
Download README.md from fulcrumresearch/echo: direct link, hf CLI and curl.
- Browser
- Download file 834 Bytes
-
https://huggingface.co/fulcrumresearch/echo/resolve/main/README.md
- Command line
-
hf download hf://fulcrumresearch/echo/README.md
-
curl -L -o README.md https://huggingface.co/fulcrumresearch/echo/resolve/main/README.md
834 Bytes
metadata
license: mit
base_model: moonshotai/Kimi-K3
base_model_relation: adapter
library_name: peft
tags:
- lora
- writing
- style-imitation
Echo
Echo is a post-trained open model that takes on the guise of a named author. It is state of the art at coherently writing in a given style.
Read more in Introducing Echo.
Model
Echo is a LoRA adapter (rank 32, alpha 32) for moonshotai/Kimi-K3, in Hugging Face PEFT format, in the model/ folder.
License
The Echo adapter is released under the MIT License. Echo is built on Kimi K3: the Kimi K3 base model and the Kimi K3 files in this repository are licensed under the Kimi K3 License, and since Echo runs on the Kimi K3 base model, its terms apply to any use of Echo.