Community to discuss about Llama, the family of large language models created by Meta AI.
openchat_3.5 openchat_3.5-16k Is the best in my opinion. I even noticed that it responds much smarter than the assistant or any bot in poe. The ones based on GPT3.
Also, it is relatively good at roleplay, although to be honest it still feels that it is not focused on it and it lacks the database to perform situations better.
No love for my Autolycus?
https://huggingface.co/FPHam/Autolycus-Mistral_7B
It's a more human language fine-tune of OpenHermes 2.5
Which are the best 30b models for RP?
Among 7B models, my recommendation for RP is https://huggingface.co/maywell/Synatra-7B-v0.3-RP. Of course, TheBloke quantized this model too.