metadata

license: gemma
library_name: mlx
pipeline_tag: text-generation
extra_gated_heading: Access Gemma on Hugging Face
extra_gated_prompt: >-
  To access Gemma on Hugging Face, you’re required to review and agree to
  Google’s usage license. To do this, please ensure you’re logged in to Hugging
  Face and click below. Requests are processed immediately.
extra_gated_button_content: Acknowledge license
base_model: google/gemma-3-1b-it
tags:
  - mlx
model-index:
  - name: gemma-3-1b-it-DQ
    results:
      - task:
          type: text-generation
        dataset:
          type: PIQA
          name: PIQA
        metrics:
          - name: pass@1
            type: pass@1
            value: 0.75
            verified: false
      - task:
          type: text-generation
        dataset:
          type: winogrande
          name: winogrande
        metrics:
          - name: pass@1
            type: pass@1
            value: 0.6
            verified: false
      - task:
          type: text-generation
        dataset:
          type: boolq
          name: boolq
        metrics:
          - name: pass@1
            type: pass@1
            value: 0.73
            verified: false
      - task:
          type: text-generation
        dataset:
          type: arc-c
          name: arc-c
        metrics:
          - name: pass@1
            type: pass@1
            value: 0.35
            verified: false

mlx-community/gemma-3-1b-it-DQ

This model mlx-community/gemma-3-1b-it-DQ was converted to MLX format from google/gemma-3-1b-it using mlx-lm version 0.25.2.

2x faster and 2.4x less memory footprint than the dequantized model

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/gemma-3-1b-it-DQ")

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)