PuchifyT1-4B-dwq6-mlx

Model Architecture and Objective

Puchify T1 integrates the S.A.F.E framework as a parallel evaluation system that operates during inference. The architecture employs specialized attention mechanisms for safety assessment while maintaining standard autoregressive generation capabilities.

This model PuchifyT1-4B-dwq6-mlx was converted to MLX format from Puchify/PuchifyT1-4B using mlx-lm version 0.26.0.

Use with mlx

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("PuchifyT1-4B-dwq6-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)
Downloads last month
13
Safetensors
Model size
1.01B params
Tensor type
BF16
·
U32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for nightmedia/PuchifyT1-4B-dwq6-mlx

Base model

Qwen/Qwen3-4B-Base
Finetuned
Qwen/Qwen3-4B
Adapter
(3)
this model

Collection including nightmedia/PuchifyT1-4B-dwq6-mlx