Model	Size	PPL	n_vocab	PPL_adjust
qwen2.5-14b-fp16.gguf	27.51	9.5316 +/- 0.08886	152064	9.5316
qwen2.5-14b-IQ4_XS.gguf	7.56	9.6508 +/- 0.09039	152064	9.6508

GGUF

Model size

14.8B params

Architecture

qwen2

4-bit

Inference API

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