Large Language Model DevOps - a package to help deploy LLMs to the Cloud.
Project description
sunholo-py
(draft release)
pip install sunholo
A python library to enable LLMOps within cloud environments
llmops
provides utilities to help manage LLM operations on Google Cloud Platform at first, but it is hoped that making it open source will help it support other clouds in the future. A lot of the functionality is not Google Cloud Platform specific, so still may be helpful.
It is derived from the Edmonbrain project, the original blog post you can read here: https://code.markedmondson.me/running-llms-on-gcp/ and owes a lot to Langchain ( https://github.com/langchain-ai/langchain )
The package includes:
agents/
- functions for working with agents, including easy flask apps, parsing chat history and dispatching requests to different agent endpointsarchive
- functions to record all Q&A activity to BigQuery via PubSubbots/
- functions for special cases regrding frontend bots such as Web Apps, Discord and Slackchunker/
- functions to slice up documents for sending into vector storescomponents/
- functions to help configure which LLM, prompt, vectorstore or document reteriver you will use based on a yaml config filedatabase/
- database setup functions and SQL to run on those sources such as Supabaseembedder/
- functions to send chunks into embedding vector storespubsub/
- use of PubSub for a message queue between componentsqna/
- utilities for running agents such as retry strats and parsing of output/inputstreaming/
- creation of streaming responses from LLM botssummarise/
- creation of summaries of large documentsutils/
- reading configuration files, Google Cloud Platform metadata
Copyright [2023] [Sunholo ApS]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
File details
Details for the file sunholo-0.1.3.20231115065331-py3-none-any.whl
.
File metadata
- Download URL: sunholo-0.1.3.20231115065331-py3-none-any.whl
- Upload date:
- Size: 61.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.18
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | fa5ce6ea991087d6efeae051dcb4c639a6f8168a7a8217151adc7257e742acd9 |
|
MD5 | dc232dd622be082bcb6f14d1d457a884 |
|
BLAKE2b-256 | 5dc247459b169e802f996c5245a723ee727603f236a2d0d4de8314ac4ea3a5d3 |