Part 4 of my local AI series is out — I gave my Ollama Discord bot a knowledge base using RAG.Stack: ChromaDB + nomic-embed-text (768-dim embeddings) + discord.py. Everything runs locally, no cloud involved.The highlight: I asked the bot who he is. He mentioned a character called Renn. I never mentioned Renn in that message. He pulled it from the right document on his own.Article: https://medium.com/@mudrastepan/gave-discord-bot-memory-building-local-ai-with-ollama-part-4-2f447c0934a6GitHub: https://github.com/StepanMudra/JakeRAG#Ollama #RAG #LocalLLM #llm #AI
Related
RE: https://helvede.net/@jwcph/117053511458535906Yeah, about that "AI watermarking" thing (that is actually more like fi...
RE: https://helvede.net/@jwcph/117053511458535906Yeah, about that "AI watermarking" thing (that is actually more like fingerprinting but words amirite...) #AI
For the past year, BayernKI – the Bavarian AI infrastructure for science – has been available for use in research at the...
For the past year, BayernKI – the Bavarian AI infrastructure for science – has been available for use in research at the LRZ and NHR@FAU.We asked computer scientist Gerhard Wellein...
#FamilyRegister Ep 8 offers us a bad translation for gender, given that the woman in the picture doesn't have a daughter...
#FamilyRegister Ep 8 offers us a bad translation for gender, given that the woman in the picture doesn't have a daughter, just 2 sons, and an unfortunate lexical choice for "confon...