The Gemma3NPC models are made with the intention to be a Roleplaying model

while the Gemma3NPC-it models are made to be embedded into a game and used as the conversation models for NPCs.


This is fintuned using the PIPPA dataset from PygmalionAI, converted into ChatML format with a custom system prompt.

Trained in Google Colab using one A100 GPU


Downloads last month
66
GGUF
Model size
6.87B params
Architecture
gemma3n
Hardware compatibility
Log In to view the estimation

8-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for chimbiwide/Gemma3NPC-Q8-GGUF

Quantized
(44)
this model

Datasets used to train chimbiwide/Gemma3NPC-Q8-GGUF

Collection including chimbiwide/Gemma3NPC-Q8-GGUF