milodev123/Mist

Mist is a conversational AI model trained on over 450,000 lines of messages from Mistium, reflecting their unique style and personality. The model excels in multi-turn dialogue but may produce subjective, opinionated responses. Its knowledge is limited to specific topics, and outputs can vary. Use it for chatbots, research, or creative writing.

How to use it

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "milodev123/Mist"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

def chat_with_mist(prompt):
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(inputs['input_ids'], max_length=150, num_beams=5, no_repeat_ngram_size=2)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

prompt = "Hey Mist, what’s your take on gaming culture?"
print(chat_with_mist(prompt))