annoy_ltm
annoy long term memory experiment for oobabooga/text-generation-webui (by YenRaven)
superbig
A prompt/context management system (by kaiokendev)
| annoy_ltm | superbig | |
|---|---|---|
| 3 | 3 | |
| 32 | 158 | |
| - | - | |
| 6.5 | 1.4 | |
| 11 months ago | about 1 year ago | |
| Python | Python | |
| - | MIT License |
The number of mentions indicates the total number of mentions that we've tracked plus the number of user suggested alternatives.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
annoy_ltm
Posts with mentions or reviews of annoy_ltm.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2023-06-18.
- Looking for the long-term memory extension.
-
I created a memory system to let your chat bots remember past interactions in a human like way.
I know very little about the difference, but I use GPTQ models so it's possible. There is one issue on the repo about something similar https://github.com/YenRaven/annoy_ltm/issues/7 If this is what you are seeing, then I think I know the issue. There is a dimensionality to the embeddings output that I kinda thought was just a magic number for most of the development, but found out there is a config on the models I use that has just the number I needed. It's possible GGML models have no such config or a different structure. I'm planing on adding a setting to the extension that will let you override this value which hopefully will fix this issue.
superbig
Posts with mentions or reviews of superbig.
We have used some of these posts to build our list of alternatives
and similar projects. The last one was on 2023-07-02.
-
Koboldcpp + Chromadb
https://github.com/kaiokendev/superbig You may want to check this or the superbooga extension for text generation webui, or sillytavern's version, and see how they implemented theirs.
-
I created a memory system to let your chat bots remember past interactions in a human like way.
There's also "SuperBIG" (more info) and an oobabooga implementation as well which can be used as LTM. (and someone is also trying to implement that into SillyTavern)
-
Is there a way to feed in documents similar to the Llama Index?
This But it looks like it's new and in early dev.
What are some alternatives?
When comparing annoy_ltm and superbig you can also consider the following projects:
long_term_memory - A gradio web UI for running Large Language Models like GPT-J 6B, OPT, GALACTICA, LLaMA, and Pygmalion.
SillyTavern - LLM Frontend for Power Users.
SillyTavern - LLM Frontend for Power Users. [Moved to: https://github.com/SillyTavern/SillyTavern]
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.
koboldcpp - A simple one-file way to run various GGML models with KoboldAI's UI
SillyTavern-Extras - Extensions API for SillyTavern.