license: gemma
base_model: Jiunsong/supergemma4-26b-abliterated-multimodal
tags:

• gemma4
• transformers
• bf16
• multimodal
• instruction-following
• tool-use
• mlx
• mlx-my-repo
  language:
• en
• ko
  pipeline_tag: text-generation
  library_name: transformers

andosen/supergemma4-26b-abliterated-multimodal-mlx-6Bit

The Model andosen/supergemma4-26b-abliterated-multimodal-mlx-6Bit was converted to MLX format from Jiunsong/supergemma4-26b-abliterated-multimodal using mlx-lm version 0.31.2.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("andosen/supergemma4-26b-abliterated-multimodal-mlx-6Bit")

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)