maciejonek/Mistral-7b-Gandalf

The Model maciejonek/Mistral-7b-Gandalf was converted to MLX format from mistralai/Mistral-7B-Instruct-v0.2 using mlx-lm version 0.21.0.

Description

Fine-tuned Mistral-7B-Instruct-v0.2 model trained for wisdom-rich prose. Trained on a dataset of philosophical musings, timeless insights, and profound reflections, it delivers responses with depth, clarity, and an almost mythical sense of guidance. Provide a concise summary or text, and Gandalf will reflect on its deeper meaning, connect it to universal truths, or add thoughtful perspectives.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("maciejonek/Mistral-7b-Gandalf")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)