Instructions to use Orvena/granite-4.2-3b-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Orvena/granite-4.2-3b-MLX-4bit with MLX:

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

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("Orvena/granite-4.2-3b-MLX-4bit")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use Orvena/granite-4.2-3b-MLX-4bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Orvena/granite-4.2-3b-MLX-4bit"

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": "Orvena/granite-4.2-3b-MLX-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use Orvena/granite-4.2-3b-MLX-4bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "Orvena/granite-4.2-3b-MLX-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "Orvena/granite-4.2-3b-MLX-4bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "Orvena/granite-4.2-3b-MLX-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use Orvena/granite-4.2-3b-MLX-4bit 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 "Orvena/granite-4.2-3b-MLX-4bit"

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 Orvena/granite-4.2-3b-MLX-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Orvena/granite-4.2-3b-MLX-4bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Orvena/granite-4.2-3b-MLX-4bit"

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 "Orvena/granite-4.2-3b-MLX-4bit" \
  --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"

granite-4.2-3b-MLX-4bit

A 4-bit MLX quantization of ibm-granite/granite-4.2-3b, converted with mlx-lm 0.31.3.

Quantization: 4 bit, group size 64 (4.501 bits per weight)
Weights: 2.06 GB safetensors
Context length: 131,072 tokens
Chat template, thinking mode, and tool calling are inherited unchanged from the base model
License: Apache 2.0, same as the base model

Use with mlx-lm

pip install mlx-lm
mlx_lm.generate --model Orvena/granite-4.2-3b-MLX-4bit --prompt "Name three Danish cities."

from mlx_lm import load, generate

model, tokenizer = load("Orvena/granite-4.2-3b-MLX-4bit")
print(generate(model, tokenizer, prompt="Name three Danish cities."))

Runs on Apple silicon Macs and recent iPhones. Peak memory during generation is about 2.2 GB.

About

Converted by the Orvena team while evaluating models for fully on-device use. See the base model card for benchmarks, training details, and intended use.

Downloads last month: -

Safetensors

Model size

0.6B params

Tensor type

BF16

U32

MLX

Hardware compatibility

4-bit

Model tree for Orvena/granite-4.2-3b-MLX-4bit

Base model

ibm-granite/granite-4.1-3b-base

Finetuned

ibm-granite/granite-4.2-3b

Quantized

(13)

this model