language:

• en
  license: apache-2.0
  tags:
• Qwen-3-14B
• instruct
• finetune
• reasoning
• hybrid-mode
• chatml
• function calling
• tool use
• json mode
• structured outputs
• atropos
• dataforge
• long context
• roleplaying
• chat
• mlx
• 3-bit
• 4-bit
• 5-bit
• 6-bit
• 8-bit
• bfloat16
• abliterated
  base_model: NousResearch/Hermes-4-14B
  library_name: mlx
  widget:
• example_title: Hermes 4
  messages:
  • role: system
    content: >-
    You are Hermes 4, a capable, neutrally-aligned assistant. Prefer concise,
    correct answers.
  • role: user
    content: Explain the difference between BFS and DFS to a new CS student.
    pipeline_tag: text-generation
    model-index:
• name: Hermes-4-Qwen-3-14B
  results: []

mlx-community/NousResearch_Hermes-4-14B-BF16-abliterated-mlx

This model mlx-community/NousResearch_Hermes-4-14B-BF16-abliterated-mlx was
converted to MLX format from n0kovo/NousResearch_Hermes-4-14B-BF16-abliterated
using mlx-lm version 0.27.1.

Quantizations are also provided in this repo in 8-, 6-, 5-, 4-, 3- and 2-bit.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/NousResearch_Hermes-4-14B-BF16-abliterated-mlx")

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)