RichardErkhov/SeacomSrl_-_SeaQwen2-0.5B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SeaQwen2-0.5B - GGUF

• Model creator: https://huggingface.co/SeacomSrl/
• Original model: https://huggingface.co/SeacomSrl/SeaQwen2-0.5B/

Name	Quant method	Size
SeaQwen2-0.5B.Q2_K.gguf	Q2_K	0.32GB
SeaQwen2-0.5B.IQ3_XS.gguf	IQ3_XS	0.32GB
SeaQwen2-0.5B.IQ3_S.gguf	IQ3_S	0.32GB
SeaQwen2-0.5B.Q3_K_S.gguf	Q3_K_S	0.32GB
SeaQwen2-0.5B.IQ3_M.gguf	IQ3_M	0.32GB
SeaQwen2-0.5B.Q3_K.gguf	Q3_K	0.33GB
SeaQwen2-0.5B.Q3_K_M.gguf	Q3_K_M	0.33GB
SeaQwen2-0.5B.Q3_K_L.gguf	Q3_K_L	0.34GB
SeaQwen2-0.5B.IQ4_XS.gguf	IQ4_XS	0.33GB
SeaQwen2-0.5B.Q4_0.gguf	Q4_0	0.33GB
SeaQwen2-0.5B.IQ4_NL.gguf	IQ4_NL	0.33GB
SeaQwen2-0.5B.Q4_K_S.gguf	Q4_K_S	0.36GB
SeaQwen2-0.5B.Q4_K.gguf	Q4_K	0.37GB
SeaQwen2-0.5B.Q4_K_M.gguf	Q4_K_M	0.37GB
SeaQwen2-0.5B.Q4_1.gguf	Q4_1	0.35GB
SeaQwen2-0.5B.Q5_0.gguf	Q5_0	0.37GB
SeaQwen2-0.5B.Q5_K_S.gguf	Q5_K_S	0.38GB
SeaQwen2-0.5B.Q5_K.gguf	Q5_K	0.39GB
SeaQwen2-0.5B.Q5_K_M.gguf	Q5_K_M	0.39GB
SeaQwen2-0.5B.Q5_1.gguf	Q5_1	0.39GB
SeaQwen2-0.5B.Q6_K.gguf	Q6_K	0.47GB
SeaQwen2-0.5B.Q8_0.gguf	Q8_0	0.49GB

Original model description:

license: apache-2.0 datasets: - SeacomSrl/rag-data language: - it

Uploaded model

• Developed by: Toti Riccardo
• License: apache-2.0
• Finetuned from model : Qwen/Qwen2-0.5B on Seacom/rag-data an italian translated dataset.

Tasks	Metric	Value		Stderr
hellaswag_it	acc	0.2945	±	0.0048
	acc_norm	0.3428	±	0.0050
arc_it	acc	0.0274	±	0.0048
	acc_norm	0.2609	±	0.0128