Allows LLMs to assist you with various tasks, all while being open-source!
Project description
DIY-Assistant
Make your own assistant through the use of LLMs
Current State
DIY Assistant is still in development, but I hope to make the app usable in a similar manner as the below code:
from helpful_assistant import Assistant
class Model:
def generate(prompt: str, stream: bool):
pass # return a generator or a string
def form_prompt(messages):
pass # not sure if we want to use. Might get in the way if using an api.
# init the app given a model class
app = Assistant(llm_class=Model, default_stream_response=True)
while True:
# Gather user input. Could be linked to transcription software, for example
user_input = input("> ")
# Allow the LLM to access the modules and actions and get their outputs
generator = app.generate(user_input, stream=True, allow_action_execution=True)
# Generate the final output of the model after the data lookups
for resp in generator:
print(resp.text, flush=True, end="")
print()
# show the data that the model accessed
print(generator.accessed_actions)
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
helpful_assistant-0.1.0.tar.gz
(18.4 kB
view details)
Built Distribution
File details
Details for the file helpful_assistant-0.1.0.tar.gz
.
File metadata
- Download URL: helpful_assistant-0.1.0.tar.gz
- Upload date:
- Size: 18.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.9.19
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 669e41aa1c0978e3faca394525e78a5bf0f5757aab5e0f657303f62b1c0c0ade |
|
MD5 | 66fb88dca201dc69cb796f25157aac17 |
|
BLAKE2b-256 | a98078a9f43651cc7cb6f3e5881166037fea7abf19dcdf0f22e63a0cd1b1d5ed |
File details
Details for the file helpful_assistant-0.1.0-py3-none-any.whl
.
File metadata
- Download URL: helpful_assistant-0.1.0-py3-none-any.whl
- Upload date:
- Size: 20.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.9.19
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6efe8cc83a1064dc7bdd85d0692e3e03feceb8e9eab16cd26884ee98b909dc17 |
|
MD5 | d4bc2e57a56f05fcb4a3374114c2dd36 |
|
BLAKE2b-256 | 070ba8ee303d8942b88acdfce8c05de4b88b1fa48daf9e86498dc776bd94184a |