Instructions to use mimaapp/sintesi-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use mimaapp/sintesi-models with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf mimaapp/sintesi-models:Q4_0
# Run inference directly in the terminal:
llama cli -hf mimaapp/sintesi-models:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf mimaapp/sintesi-models:Q4_0
# Run inference directly in the terminal:
llama cli -hf mimaapp/sintesi-models:Q4_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf mimaapp/sintesi-models:Q4_0
# Run inference directly in the terminal:
./llama-cli -hf mimaapp/sintesi-models:Q4_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf mimaapp/sintesi-models:Q4_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mimaapp/sintesi-models:Q4_0

Use Docker

docker model run hf.co/mimaapp/sintesi-models:Q4_0

LM Studio
Jan
Ollama
How to use mimaapp/sintesi-models with Ollama:
```
ollama run hf.co/mimaapp/sintesi-models:Q4_0
```
Unsloth Desktop

How to use mimaapp/sintesi-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mimaapp/sintesi-models:Q4_0

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "mimaapp/sintesi-models:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use mimaapp/sintesi-models with Docker Model Runner:
```
docker model run hf.co/mimaapp/sintesi-models:Q4_0
```

Lemonade

How to use mimaapp/sintesi-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mimaapp/sintesi-models:Q4_0

Run and chat with the model

lemonade run user.sintesi-models-Q4_0

List all available models

lemonade list

Hermes Agent

How to use mimaapp/sintesi-models with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mimaapp/sintesi-models:Q4_0

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 mimaapp/sintesi-models:Q4_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use mimaapp/sintesi-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mimaapp/sintesi-models:Q4_0

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 "mimaapp/sintesi-models:Q4_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Sintesi — on-device models

Mirror of the three model files downloaded by the Sintesi Android app (com.mimaapp.sintesi) to transcribe and rewrite voice notes on the phone, without sending recordings anywhere. The files are byte-identical copies of the originals; they are hosted here only so the app does not depend on someone else's repository.

Each file keeps its own license. There is no single license for the whole repository.

File	Original	License
`ggml-parakeet-tdt-0.6b-v3-q8_0.bin`	ggml-org/parakeet-GGUF, a q8_0 conversion of nvidia/parakeet-tdt-0.6b-v3	CC-BY-4.0
`ggml-silero-v5.1.2.bin`	ggml-org/whisper-vad, a conversion of Silero VAD v5.1.2	MIT
`Qwen3-4B-Instruct-2507-Q4_0.gguf`	unsloth/Qwen3-4B-Instruct-2507-GGUF, a Q4_0 quantization of Qwen/Qwen3-4B-Instruct-2507	Apache 2.0

Attribution

Parakeet TDT 0.6B v3 by NVIDIA, licensed under CC-BY-4.0. Changes: converted and quantized (q8_0) to the ggml format by ggml-org; no further changes.
Silero VAD by Silero Team, licensed under MIT (copyright notice and license text in LICENSE-Silero-VAD-MIT.txt). Converted to the ggml format by ggml-org; no further changes.
Qwen3-4B-Instruct-2507 by the Qwen team, Alibaba Cloud, licensed under Apache 2.0 (license text in LICENSE-Qwen3-Apache-2.0.txt). Changes: quantized (Q4_0) to the GGUF format by Unsloth; no further changes.

Checksums (SHA-256)

4d64e9e96c2792186d072fde0034df0ad670cf680a2f53069052ead827fd600e  ggml-parakeet-tdt-0.6b-v3-q8_0.bin
29940d98d42b91fbd05ce489f3ecf7c72f0a42f027e4875919a28fb4c04ea2cf  ggml-silero-v5.1.2.bin
e0ba675d86ab277c61701c6793659b2ae801d95e3be791464c321e6fbf613be2  Qwen3-4B-Instruct-2507-Q4_0.gguf

Downloads last month: -

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support