Quantization made by Richard Erkhov.

Llama2_init_Mistral - GGUF

Model creator: https://huggingface.co/Cartinoe5930/
Original model: https://huggingface.co/Cartinoe5930/Llama2_init_Mistral/

Name	Quant method	Size
Llama2_init_Mistral.Q2_K.gguf	Q2_K	2.53GB
Llama2_init_Mistral.IQ3_XS.gguf	IQ3_XS	2.81GB
Llama2_init_Mistral.IQ3_S.gguf	IQ3_S	2.96GB
Llama2_init_Mistral.Q3_K_S.gguf	Q3_K_S	2.95GB
Llama2_init_Mistral.IQ3_M.gguf	IQ3_M	3.06GB
Llama2_init_Mistral.Q3_K.gguf	Q3_K	3.28GB
Llama2_init_Mistral.Q3_K_M.gguf	Q3_K_M	3.28GB
Llama2_init_Mistral.Q3_K_L.gguf	Q3_K_L	3.56GB
Llama2_init_Mistral.IQ4_XS.gguf	IQ4_XS	3.67GB
Llama2_init_Mistral.Q4_0.gguf	Q4_0	3.83GB
Llama2_init_Mistral.IQ4_NL.gguf	IQ4_NL	3.87GB
Llama2_init_Mistral.Q4_K_S.gguf	Q4_K_S	3.86GB
Llama2_init_Mistral.Q4_K.gguf	Q4_K	4.07GB
Llama2_init_Mistral.Q4_K_M.gguf	Q4_K_M	4.07GB
Llama2_init_Mistral.Q4_1.gguf	Q4_1	4.24GB
Llama2_init_Mistral.Q5_0.gguf	Q5_0	4.65GB
Llama2_init_Mistral.Q5_K_S.gguf	Q5_K_S	4.65GB
Llama2_init_Mistral.Q5_K.gguf	Q5_K	4.78GB
Llama2_init_Mistral.Q5_K_M.gguf	Q5_K_M	4.78GB
Llama2_init_Mistral.Q5_1.gguf	Q5_1	5.07GB
Llama2_init_Mistral.Q6_K.gguf	Q6_K	5.53GB
Llama2_init_Mistral.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0

Base Model - SOLAR-10.7B

This model is the base model implementation of SOLAR-10.7B. The architecture of base model is Llama2 architecture and initialized with weights of Mistral.

Please check specific details in the GitHub Repository.

GitHub Repository: https://github.com/gauss5930/iDUS

🏆 HuggingFace Open LLM Leaderboard

Model	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K	Average
Llama2_init_Mistral	60.07	83.3	64.09	42.15	78.37	37.91	60.98