Find answers from the community

Updated 2 years ago

Yes a chat interface like the ChatGPT

Yes, a chat interface like the ChatGPT chatbot. Not software development, just use for academic purposes.
L
S
s
5 comments
Assuming you have the hardware to run the model, I always check the leaderboard https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
You'll need a beefy GPU though.

You can also look into using llama.cpp if you are stuck with CPU

https://github.com/ggerganov/llama.cpp

Plain Text
from llama_index import LLMPredictor, ServiceContext
from langchain.llms import LlamaCpp

llm = LlamaCpp(model_path="./ggml-model-q4_0.bin")

llm_predictor = LLMPredictor(llm=llm)

service_context = ServiceContext.from_defaults(llm_predictor=llm_predictor, chunk_size=512, context_window=<??>)


(I forget the input size to these models lol)
Is there any blog or tutorial on how to download/install these models on my PC?
Sounds like sailing, which is described as "the most expensive way to get anywhere for free"
I ask because there are Open Source models such as Vicuna, Alpaca, Koala, etc. I imagine that there could be free chatbots sites based on these models.
Add a reply
Sign up and join the conversation on Discord