RichardErkhov/FelixChao_-_NarutoDolphin-10B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

NarutoDolphin-10B - GGUF

• Model creator: https://huggingface.co/FelixChao/
• Original model: https://huggingface.co/FelixChao/NarutoDolphin-10B/

Name	Quant method	Size
NarutoDolphin-10B.Q2_K.gguf	Q2_K	3.73GB
NarutoDolphin-10B.IQ3_XS.gguf	IQ3_XS	4.14GB
NarutoDolphin-10B.IQ3_S.gguf	IQ3_S	4.37GB
NarutoDolphin-10B.Q3_K_S.gguf	Q3_K_S	4.34GB
NarutoDolphin-10B.IQ3_M.gguf	IQ3_M	4.51GB
NarutoDolphin-10B.Q3_K.gguf	Q3_K	4.84GB
NarutoDolphin-10B.Q3_K_M.gguf	Q3_K_M	4.84GB
NarutoDolphin-10B.Q3_K_L.gguf	Q3_K_L	5.26GB
NarutoDolphin-10B.IQ4_XS.gguf	IQ4_XS	5.43GB
NarutoDolphin-10B.Q4_0.gguf	Q4_0	5.66GB
NarutoDolphin-10B.IQ4_NL.gguf	IQ4_NL	5.72GB
NarutoDolphin-10B.Q4_K_S.gguf	Q4_K_S	5.7GB
NarutoDolphin-10B.Q4_K.gguf	Q4_K	6.02GB
NarutoDolphin-10B.Q4_K_M.gguf	Q4_K_M	6.02GB
NarutoDolphin-10B.Q4_1.gguf	Q4_1	6.27GB
NarutoDolphin-10B.Q5_0.gguf	Q5_0	6.89GB
NarutoDolphin-10B.Q5_K_S.gguf	Q5_K_S	6.89GB
NarutoDolphin-10B.Q5_K.gguf	Q5_K	7.08GB
NarutoDolphin-10B.Q5_K_M.gguf	Q5_K_M	7.08GB
NarutoDolphin-10B.Q5_1.gguf	Q5_1	7.51GB
NarutoDolphin-10B.Q6_K.gguf	Q6_K	8.2GB
NarutoDolphin-10B.Q8_0.gguf	Q8_0	10.62GB

Original model description:

license: apache-2.0 tags: - merge - FelixChao/WizardDolphin-7B - FelixChao/NinjaDolphin-7B

NarutoDolphin-10B

NarutoDolphin-10B is a merge of the following models:

• FelixChao/WizardDolphin-7B
• FelixChao/NinjaDolphin-7B

Quantizationed version

Quantizationed version of this model is available thanks to s3nh.

GGUF

• s3nh/NarutoDolphin-10B-GGUF

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "FelixChao/NarutoDolphin-10B"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])