#Wanting tips for local llms

1 messages · Page 1 of 1 (latest)

raven pagoda
#

I’m making a dual llm assistant with Llama 3.1 (third-gen Llama) and Phi-3 (third-gen Phi) and was just wanting some tips. Im using the chat gpt and Claude ai as a was to translate what I want into code but if anyone knows of better models or a dedicated text to code model that is local all is welcome. Please dm me with any info.

#

I’m also wanting to integrate in a TTS and a STT model as well for communication verbally

raven pagoda
#

update switched to only one ai, was having problems with personality routing

simple quartz
#

The deep-seek coder models are good in my opinion.