Announcing AnythingLLM. An open-source (MIT) full-stack app for chatting with... anything.
β UI for managing documents
β Uses @OpenAI , @pinecone & @LangChainAI
π« massive RAM requirements
π« LLM downloading or training
π« paying to re-embed documents
+ data tooling! twitter.com/i/web/status/1β¦
β UI for managing documents
β Uses @OpenAI , @pinecone & @LangChainAI
π« massive RAM requirements
π« LLM downloading or training
π« paying to re-embed documents
+ data tooling! twitter.com/i/web/status/1β¦
First, you can find AnythingLLM on Github here
github.com/Mintplex-Labs/β¦
Requirements:
- Python 3.8+
- npm and Node v16+
Thats it. No bulky CPU or GPU is required.
Chatting with your documents is the "hello world" of LLM use-cases, why not make it more accessible?
github.com/Mintplex-Labs/β¦
Requirements:
- Python 3.8+
- npm and Node v16+
Thats it. No bulky CPU or GPU is required.
Chatting with your documents is the "hello world" of LLM use-cases, why not make it more accessible?
What is so special about AnythingLLM?
π₯ No crazy system requirements - runs fast and passively on your machine
π§° Full data-collection tool suite. Collect anything
π Entire YouTube Channels
π Substacks
π Mediums
π Gitbooks
+ local document processing
π₯ No crazy system requirements - runs fast and passively on your machine
π§° Full data-collection tool suite. Collect anything
π Entire YouTube Channels
π Substacks
π Mediums
π Gitbooks
+ local document processing
β‘ Efficient vector-caching
Embed forever, pay once.
Embedding is low-cost using OpenAIs ada text embedding, but over thousands of tokens - why pay more than once?
AnythingLLMs vector caching allows you to use the same document across workspaces and pay once.
Embed forever, pay once.
Embedding is low-cost using OpenAIs ada text embedding, but over thousands of tokens - why pay more than once?
AnythingLLMs vector caching allows you to use the same document across workspaces and pay once.
π Persistent and Useable
Chatting and managing documents is all done from a web-browser UI.
Shut down the app and start it later - all your documents, chats, and more are still present. Picking up right where you left off.
The database is locally saved on your machine.
Chatting and managing documents is all done from a web-browser UI.
Shut down the app and start it later - all your documents, chats, and more are still present. Picking up right where you left off.
The database is locally saved on your machine.
π Bring-Your-Own-Keys
Use your own OpenAI API key and Pinecone (free) instance and you are already on the way to managing documents with ease.
This keeps overhead low by just using what is existing, easy and accessible.
Use your own OpenAI API key and Pinecone (free) instance and you are already on the way to managing documents with ease.
This keeps overhead low by just using what is existing, easy and accessible.
AnythingLLM is MIT Licensed so go crazy. Read more about AnythingLLM on Medium at:
medium.com/rampp/anythingβ¦
medium.com/rampp/anythingβ¦
Just merged in @trychroma for all the Chroma-homies who are down to use more OSS tools
github.com/Mintplex-Labs/β¦
github.com/Mintplex-Labs/β¦
β’ β’ β’
Missing some Tweet in this thread? You can try to
force a refresh
γ
