base_model: nicoboss/Hermes-3-Llama-3.1-405B-Uncensored
library_name: peft
tags:

• generated_from_trainer
• mlx
• mlx-my-repo
  license: llama3.1
  datasets:
• Guilherme34/uncensor
  model-index:
• name: Hermes-3-Llama-3.1-405B-Uncensored
  results: []

Dessicated/Hermes-3-Llama-3.1-405B-Uncensored-Q2-mlx

The Model Dessicated/Hermes-3-Llama-3.1-405B-Uncensored-Q2-mlx was converted to MLX format from nicoboss/Hermes-3-Llama-3.1-405B-Uncensored using mlx-lm version 0.21.5.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("Dessicated/Hermes-3-Llama-3.1-405B-Uncensored-Q2-mlx")

prompt="hello"

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

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