Haary/haryra-7b-mistral-id

Haary/haryra-7b-mistral-id is QLoRA quantized version of Ichsan2895/Merak-7B-v4

Install the necessary packages

Requires: Transformers from source - only needed for versions <= v4.34.

# Install transformers from source - only needed for versions <= v4.34
!pip install git+https://github.com/huggingface/transformers.git
!pip install accelerate

Example Python code

import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="Haary/haryra-7b-mistral-id", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "Anda adalah chatbot ramah yang selalu merespons dengan singkat dan jelas",
    },
    {"role": "user", "content": "Apa bedanya antara raspberry pi dan esp32?"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Credits

Ichsan2895/Merak-7B-v4 for base model.

image source

pixabay.com