Modelpublic

RichardErkhov/FunDialogues_-_llamav2-LoRaco-7b-merged-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes422downloads

Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llamav2-LoRaco-7b-merged - GGUF

—Model creator: https://huggingface.co/FunDialogues/
—Original model: https://huggingface.co/FunDialogues/llamav2-LoRaco-7b-merged/

Name	Quant method	Size
llamav2-LoRaco-7b-merged.Q2_K.gguf	Q2_K	2.36GB
llamav2-LoRaco-7b-merged.IQ3_XS.gguf	IQ3_XS	2.6GB
llamav2-LoRaco-7b-merged.IQ3_S.gguf	IQ3_S	2.75GB
llamav2-LoRaco-7b-merged.Q3_K_S.gguf	Q3KS	2.75GB
llamav2-LoRaco-7b-merged.IQ3_M.gguf	IQ3_M	2.9GB
llamav2-LoRaco-7b-merged.Q3_K.gguf	Q3_K	3.07GB
llamav2-LoRaco-7b-merged.Q3_K_M.gguf	Q3KM	3.07GB
llamav2-LoRaco-7b-merged.Q3_K_L.gguf	Q3KL	3.35GB
llamav2-LoRaco-7b-merged.IQ4_XS.gguf	IQ4_XS	3.4GB
llamav2-LoRaco-7b-merged.Q4_0.gguf	Q4_0	3.56GB
llamav2-LoRaco-7b-merged.IQ4_NL.gguf	IQ4_NL	3.58GB
llamav2-LoRaco-7b-merged.Q4_K_S.gguf	Q4KS	3.59GB
llamav2-LoRaco-7b-merged.Q4_K.gguf	Q4_K	3.8GB
llamav2-LoRaco-7b-merged.Q4_K_M.gguf	Q4KM	3.8GB
llamav2-LoRaco-7b-merged.Q4_1.gguf	Q4_1	3.95GB
llamav2-LoRaco-7b-merged.Q5_0.gguf	Q5_0	4.33GB
llamav2-LoRaco-7b-merged.Q5_K_S.gguf	Q5KS	4.33GB
llamav2-LoRaco-7b-merged.Q5_K.gguf	Q5_K	4.45GB
llamav2-LoRaco-7b-merged.Q5_K_M.gguf	Q5KM	4.45GB
llamav2-LoRaco-7b-merged.Q5_1.gguf	Q5_1	4.72GB
llamav2-LoRaco-7b-merged.Q6_K.gguf	Q6_K	5.15GB
llamav2-LoRaco-7b-merged.Q8_0.gguf	Q8_0	6.67GB

Original model description: --- library_name: transformers license: apache-2.0 datasets:

—timdettmers/openassistant-guanaco pipeline_tag: text-generation language:
—en
—es
—ru
—de
—zh
—fr
—th
—pt
—ca
—ko
—uk
—it
—ja tags:
—gaudi
—intel
—lora ---

Model Card for Model ID

This model is merge of meta-llama/Llama-2-7b-hf and FunDialogues/llamav2-LoRaco-7b. FunDialogues/llamav2-LoRaco-7b adapter weights were finetuned using the timdettmers/openassistant-guanaco dataset.