KillianLucas / ooba

Run language models on consumer hardware.
GNU Affero General Public License v3.0
26 stars 6 forks source link

Open In Colab

A headless Oobabooga wrapper.


pip install ooba
import ooba

path = ooba.download("https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF")

llm = ooba.llm(path)
messages = [{"role": "user", "content": "Hi Mistral."}]

for token in llm.chat(messages):
    print(token)


https://github.com/KillianLucas/ooba/assets/63927363/b741f4ee-9dca-4c50-9405-dd99550b2dc8