News for hackers. Stuff that matters.

tankado

Re:Run llama3 locally with 1M token context on Wednesday May 01, 2024 @03:28AM (#40220479)
"Note: using a 256k context window requires at least 64GB of memory. Using a 1M+ context window requires significantly more (100GB+)."
The Cheshire Cat is an open source, customizable AI architecture: - language model agnostic (works with OpenAI, Cohere, HuggingFace models, custom)- long term memory - can use external tools (APIs, other models)- can ingest documents (.pdf, .txt) - 100% dockerized- extendible via plugins

Waiting for you to try it out and contribute with tutorials, code, and whatever makes you happy

GitHub Repo: https://github.com/pieroit/cheshire-cat

Re:I tracked everything I read on the internet for a year on Thursday September 08, 2022 @12:27PM (#32767881)
I tried something similar but I failed after a couple of months
Re:I tracked everything I read on the internet for a year on Thursday September 08, 2022 @12:25PM (#32767866)
I have a similar approach but I use Instapaper