Instructions to use mlx-community/diffusiongemma-26B-A4B-it-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/diffusiongemma-26B-A4B-it-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("mlx-community/diffusiongemma-26B-A4B-it-8bit")
config = load_config("mlx-community/diffusiongemma-26B-A4B-it-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 mlx-community/diffusiongemma-26B-A4B-it-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 "mlx-community/diffusiongemma-26B-A4B-it-8bit"

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": "mlx-community/diffusiongemma-26B-A4B-it-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mlx-community/diffusiongemma-26B-A4B-it-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 "mlx-community/diffusiongemma-26B-A4B-it-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 mlx-community/diffusiongemma-26B-A4B-it-8bit

Run Hermes

hermes

OpenClaw new

How to use mlx-community/diffusiongemma-26B-A4B-it-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 "mlx-community/diffusiongemma-26B-A4B-it-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 "mlx-community/diffusiongemma-26B-A4B-it-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"

Are these models good for coding agents?

by wsintra - opened 29 days ago

Discussion

wsintra

29 days ago

Tried running in omlx with opencode.

omlx.server - WARNING - POST /v1/chat/completions -> 400: Tool calling is not supported with diffusion models.

datagram

MLX Community org 29 days ago

•

edited 29 days ago

Optimally, a diffusion model would not be the best choice for competitive business GTM or even maintenance engineering. But it could probably build a snake game for your github...idk...could be good for pasting images and asking for code to build out something similar or not to the image.

/> GGUF is basically a wrapper that compresses AI models (mainly text, but sometimes visual models) so you can run them with less computer memory.
/> Diffusion is the AI engine that paints pixels or generates frames.
/> Quantization is the act of shrinking these engines so they fit on your personal hardware (i.e., mlx apple silicon, yada yada yada).

https://huggingface.co/blog/ngxson/common-ai-model-formats

asking the questions means you are on the right track!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment