Instructions to use xav-ie/scribe-cleanup-mlx-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xav-ie/scribe-cleanup-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("xav-ie/scribe-cleanup-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 xav-ie/scribe-cleanup-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 "xav-ie/scribe-cleanup-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": "xav-ie/scribe-cleanup-mlx-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use xav-ie/scribe-cleanup-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 "xav-ie/scribe-cleanup-mlx-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "xav-ie/scribe-cleanup-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": "xav-ie/scribe-cleanup-mlx-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use xav-ie/scribe-cleanup-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 "xav-ie/scribe-cleanup-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 xav-ie/scribe-cleanup-mlx-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use xav-ie/scribe-cleanup-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 "xav-ie/scribe-cleanup-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 "xav-ie/scribe-cleanup-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"

scribe cleanup v1 (Macs with Apple chips)

A small model that tidies meeting transcripts. It removes "um", stutters and false starts, keeps only the final value when someone corrects themselves, and fixes punctuation and capitals. What was said, and every number, stays the same. This is the compressed version for Macs with Apple chips, and it scores the same as the full model on our tests.

Results, limits and how it was built are on the main card: xav-ie/scribe-cleanup.

Version	For	Size
`xav-ie/scribe-cleanup-mlx-4bit`	Macs with Apple chips (compressed)	2.8 GB
`xav-ie/scribe-cleanup-GGUF`	llama.cpp, a popular tool for running models locally (compressed)	3.3 GB
`xav-ie/scribe-cleanup`	the full-size model, for Python's `transformers` library or vLLM on a server	9.3 GB
`xav-ie/scribe-cleanup-lora`	just the trained add-on (LoRA), for building your own versions	93 MB

Quick start

pip install mlx-lm
mlx_lm.server --model xav-ie/scribe-cleanup-mlx-4bit --port 8080

Or, about 1.5 times faster at the same quality thanks to prompt lookup, serve_mlx.py (needs uv): ./serve_mlx.py --port 8080.

Send prompt.txt as the system message and the lines inside <transcript> tags, with temperature 0, then run the result through guard.py, the safety check that keeps the original line wherever a number, a "not", a speaker or a timestamp changed. The main card's "How to use" has details and a Python example.

Files

File
`model.safetensors` and config	4-bit MLX (group size 32), 2.8 GB
`prompt.txt`	the instructions the model was trained with
`guard.py`	the safety check (plain Python)
`serve_mlx.py`	a server for this model with prompt lookup

License: Apache-2.0 (LICENSE), built on google/gemma-4-E2B-it (Apache-2.0).

Downloads last month: 19

Safetensors

Model size

5B params

Tensor type

U32

BF16

MLX

Hardware compatibility

4-bit

Model tree for xav-ie/scribe-cleanup-mlx-4bit

Base model

google/gemma-4-E2B

Finetuned

google/gemma-4-E2B-it

Finetuned

xav-ie/scribe-cleanup

Quantized

(2)

this model