You can deploy this model and start earning money today!

RichardErkhov/
inswave_-_AISquare-Instruct-llama2-koen-13b-v0.9.24-gguf

Name: inswave_-_AISquare-Instruct-llama2-koen-13b-v0.9.24-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

AISquare-Instruct-llama2-koen-13b-v0.9.24 - GGUF

Model creator: https://huggingface.co/inswave/
Original model: https://huggingface.co/inswave/AISquare-Instruct-llama2-koen-13b-v0.9.24/

Name	Quant method	Size
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q2_K.gguf	Q2_K	4.6GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.IQ3_XS.gguf	IQ3_XS	5.08GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.IQ3_S.gguf	IQ3_S	5.36GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q3_K_S.gguf	Q3_K_S	5.36GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.IQ3_M.gguf	IQ3_M	5.66GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q3_K.gguf	Q3_K	5.99GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q3_K_M.gguf	Q3_K_M	5.99GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q3_K_L.gguf	Q3_K_L	6.54GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.IQ4_XS.gguf	IQ4_XS	6.63GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q4_0.gguf	Q4_0	6.95GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.IQ4_NL.gguf	IQ4_NL	7.0GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q4_K_S.gguf	Q4_K_S	7.01GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q4_K.gguf	Q4_K	7.42GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q4_K_M.gguf	Q4_K_M	7.42GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q4_1.gguf	Q4_1	7.71GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q5_0.gguf	Q5_0	8.46GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q5_K_S.gguf	Q5_K_S	8.46GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q5_K.gguf	Q5_K	8.7GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q5_K_M.gguf	Q5_K_M	8.7GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q5_1.gguf	Q5_1	9.21GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q6_K.gguf	Q6_K	10.06GB
AISquare-Instruct-llama2-koen-13b-v0.9.24.Q8_0.gguf	Q8_0	13.03GB

Original model description:

language:

en pipeline_tag: text-generation license: cc-by-nc-4.0

AISquare-Instruct-llama2-koen-13b-v0.9.24

Model Details

Developed by Inswave Systems UI Platform Team

Method
Using DPO method and SFT method

Hardware
We utilized an A100x4 * 1 for training our model

Base Model beomi/llama2-koen-13b

Implementation Code

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "inswave/AISquare-Instruct-llama2-koen-13b-v0.9.24"
model = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
tokenizer = AutoTokenizer.from_pretrained(repo)

RichardErkhov/inswave_-_AISquare-Instruct-llama2-koen-13b-v0.9.24-gguf

Original model description:

AISquare-Instruct-llama2-koen-13b-v0.9.24

Model Details

Implementation Code

RichardErkhov/
inswave_-_AISquare-Instruct-llama2-koen-13b-v0.9.24-gguf