RESMPDEV/Qwen1.5-Wukong-0.5B

RESMPDEV/Qwen1.5-Wukong-0.5B is a 0.6 billion parameter, 32K context length, decoder-only language model. It is a chat finetune of the Qwen1.5-0.5B base model, specifically dealigned and trained on the Teknium OpenHermes-2.5 dataset and supplementary data from Cognitive Computations. This model is designed for chat applications, offering a specialized alternative to the original Qwen1.5 series.

Warm
Public
0.6B
BF16
32768
License: tongyi-qianwen-research
Hugging Face

No reviews yet. Be the first to review!