LMCocktail-phi-2-v1-GGUF/README.md

-license, -license_link, -license_name, base_model, inference, model_creator, model_name, pipeline_tag, quantized_by, tags

-license	-license_link	-license_name	base_model	inference	model_creator	model_name	pipeline_tag	quantized_by	tags
other	https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE	microsoft-research-license	Yhyu13/LMCocktail-phi-2-v1	false	Yhyu13	LMCocktail-phi-2-v1	text-generation	afrideva	gguf ggml quantized q2_k q3_k_m q4_k_m q5_k_m q6_k q8_0

Yhyu13/LMCocktail-phi-2-v1-GGUF

Quantized GGUF model files for LMCocktail-phi-2-v1 from Yhyu13

Name	Quant method	Size
lmcocktail-phi-2-v1.fp16.gguf	fp16	5.56 GB
lmcocktail-phi-2-v1.q2_k.gguf	q2_k	1.17 GB
lmcocktail-phi-2-v1.q3_k_m.gguf	q3_k_m	1.48 GB
lmcocktail-phi-2-v1.q4_k_m.gguf	q4_k_m	1.79 GB
lmcocktail-phi-2-v1.q5_k_m.gguf	q5_k_m	2.07 GB
lmcocktail-phi-2-v1.q6_k.gguf	q6_k	2.29 GB
lmcocktail-phi-2-v1.q8_0.gguf	q8_0	2.96 GB

Original Model Card:

LM-cocktail phi-2 v1

This is a 50%-50% model of the phi2 alpca gpt4 and phi2 ultrachat200K models.

https://huggingface.co/Yhyu13/phi-2-sft-alpaca_gpt4_en-ep1

https://huggingface.co/venkycs/phi-2-ultrachat200k

Code

The LM-cocktail is novel technique for merging multiple models https://arxiv.org/abs/2311.13534

Code is backed up by this repo https://github.com/FlagOpen/FlagEmbedding.git

Merging scripts available under the ./scripts folder