Instructions to use unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 with MLX:

# Make sure mlx-vlm is installed
# pip install --upgrade mlx-vlm

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

# Load the model
model, processor = load("unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8")
config = load_config("unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8")

# Prepare input
image = ["http://images.cocodataset.org/val2017/000000039769.jpg"]
prompt = "Describe this image."

# Apply chat template
formatted_prompt = apply_chat_template(
    processor, config, prompt, num_images=1
)

# Generate output
output = generate(model, processor, formatted_prompt, image)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8

An MLX MXFP8 quantization (microscaling FP8, group_size = 32) of microsoft/GELab-Zero-4B-preview-Sico-Evolution, a Qwen3-VL vision-language model, converted from the original bf16 weights for Apple Silicon.

Base model: microsoft/GELab-Zero-4B-preview-Sico-Evolution
Architecture: Qwen3VLForConditionalGeneration (qwen3_vl) — vision-language
Quantization: MLX MXFP8 — mode=mxfp8, 8-bit, group size 32 (~8.98 bits/weight incl. block scales)
Size on disk: ~4.7 GB

MXFP8 keeps a per-32-element E8M0 microscale, which better preserves dynamic range than affine int8 at a similar footprint.

Use with mlx-vlm

pip install mlx-vlm
python -m mlx_vlm generate \
  --model unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8 \
  --prompt "Describe this image." \
  --image path/to/image.png

Text-only prompting works as well.

License

Inherits the base model's Apache-2.0 license.

Downloads last month: 14

Safetensors

Model size

2B params

Tensor type

U32

BF16

MLX

Hardware compatibility

8-bit

Model tree for unigilby/GELab-Zero-4B-preview-Sico-Evolution-MLX-mxfp8

Base model

microsoft/GELab-Zero-4B-preview-Sico-Evolution

Quantized

(5)

this model