Instructions to use beingpax/VoiceInk-Refine-V1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use beingpax/VoiceInk-Refine-V1 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("beingpax/VoiceInk-Refine-V1")

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 beingpax/VoiceInk-Refine-V1 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "beingpax/VoiceInk-Refine-V1"

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": "beingpax/VoiceInk-Refine-V1"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use beingpax/VoiceInk-Refine-V1 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 "beingpax/VoiceInk-Refine-V1"

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 beingpax/VoiceInk-Refine-V1

Run Hermes

hermes

OpenClaw new

How to use beingpax/VoiceInk-Refine-V1 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "beingpax/VoiceInk-Refine-V1"

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 "beingpax/VoiceInk-Refine-V1" \
  --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"

MLX LM

How to use beingpax/VoiceInk-Refine-V1 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "beingpax/VoiceInk-Refine-V1"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "beingpax/VoiceInk-Refine-V1"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "beingpax/VoiceInk-Refine-V1",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

VoiceInk Refine V1

A compact cleanup model for VoiceInk. It turns raw speech-to-text output into clear, ready-to-use writing while keeping the speaker's meaning and tone.

It handles punctuation, capitalization, filler words, repetitions, spoken formatting cues, lists, and email-style paragraphs.

Model

VoiceInk Refine V1 is a task-specific fine-tune of Qwen3.5-2B for ASR transcript refinement. This release uses 4-bit post-training weight quantization.

Example

Input

hi maya comma quick update period new paragraph can we meet on friday question mark thanks comma daniel

Output

Hi Maya,

Quick update.

Can we meet on Friday?

Thanks,
Daniel

License

This model is provided for use with VoiceInk. See LICENSE.md for details.

Downloads last month: 34

Safetensors

Model size

0.3B params

Tensor type

BF16

U32

F32

MLX

Hardware compatibility

4-bit