Instructions to use nativ-community/GEV-26B-Decide-MLX-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nativ-community/GEV-26B-Decide-MLX-8bit 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("nativ-community/GEV-26B-Decide-MLX-8bit")
config = load_config("nativ-community/GEV-26B-Decide-MLX-8bit")

# 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 nativ-community/GEV-26B-Decide-MLX-8bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "nativ-community/GEV-26B-Decide-MLX-8bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "nativ-community/GEV-26B-Decide-MLX-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use nativ-community/GEV-26B-Decide-MLX-8bit 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 "nativ-community/GEV-26B-Decide-MLX-8bit"

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 nativ-community/GEV-26B-Decide-MLX-8bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use nativ-community/GEV-26B-Decide-MLX-8bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "nativ-community/GEV-26B-Decide-MLX-8bit"

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 "nativ-community/GEV-26B-Decide-MLX-8bit" \
  --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"

GEV-26B-Decide-MLX-8bit

MLX conversion of autotrust/GEV-26B-Decide for mlx-vlm, with the System 1 LoRA merged and the decision head kept in float32. GEV is a decision model: it returns a probability for every option of a question in one forward pass and does not generate text.

GEV support is not yet in an mlx-vlm release:

pip install "git+https://github.com/Lazarus-931/mlx-vlm.git@feat/gev"

from mlx_vlm import load, predict

model, processor = load("nativ-community/GEV-26B-Decide-MLX-8bit")
result = predict(model, processor, "The parcel arrived damaged and I want my money back.", {
    "refund": {"type": "bool", "instructions": "Is the customer asking for a refund?"},
})
print(result["answers"]["refund"]["value"])

Field	Value
Source	`autotrust/GEV-26B-Decide`
Source revision	`7c89590ead085bf77630b4bf68264ea30b6ddc78`
Quantization	`affine 8-bit, group size 64`
mlx-vlm	`Lazarus-931/mlx-vlm@feat/gev`
Verification	Same answer as autotrust's transformers + peft System 1 reference on 7/7 test questions (bool, choice, score, JSON state, 20-option tournament, image, two images); identical token ids on 9/9 forward passes; largest probability gap 0.018

Downloads last month: 35

Safetensors

Model size

26B params

Tensor type

BF16

F32

U32

MLX

Hardware compatibility

8-bit

Model tree for nativ-community/GEV-26B-Decide-MLX-8bit

Base model

google/gemma-4-26B-A4B

Finetuned

google/gemma-4-26B-A4B-it

Adapter

autotrust/GEV-26B-Decide

Quantized

(8)

this model