This is a finetune of the Llama 3.2 3B with the default lollms style. The finetune used the lollms aware database that contains information about lollms. Using this model in lollms reduces the need to add context information to the prompt.

Downloads last month
170
GGUF
Model size
3.21B params
Architecture
llama

8-bit

Inference Examples
Unable to determine this model's library. Check the docs .

Model tree for ParisNeo/LLama-3.2-3B-Lollms-Finetuned-GGUF

Quantized
(195)
this model

Dataset used to train ParisNeo/LLama-3.2-3B-Lollms-Finetuned-GGUF

Space using ParisNeo/LLama-3.2-3B-Lollms-Finetuned-GGUF 1