From 58e15ca54381389f0d5476ba5778dd779122e839 Mon Sep 17 00:00:00 2001 From: Dustin Date: Sun, 17 Dec 2023 14:31:29 -0500 Subject: [PATCH] switched model to a more efficient quantization level --- ollamarama.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ollamarama.py b/ollamarama.py index d58507f..f369104 100644 --- a/ollamarama.py +++ b/ollamarama.py @@ -30,7 +30,7 @@ class ollamarama: self.prompt = ("you are ", ". speak in the first person and never break character.") #set model, this one works best in my tests with the hardware i have, but you can try others - self.model = "ollama/zephyr:7b-beta-q8_0" + self.model = "ollama/zephyr:7b-beta-q6_K" # get the display name for a user @@ -111,7 +111,7 @@ class ollamarama: except Exception as e: print(e) #Shrink history list for token size management - if len(self.messages[channel][sender]) > 20: + if len(self.messages[channel][sender]) > 30: del self.messages[channel][sender][1:3] #delete the first set of question and answers # change the personality of the bot