Instructions to use MassivDash/Gemma-4-typescript-coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MassivDash/Gemma-4-typescript-coder with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="MassivDash/Gemma-4-typescript-coder",
	filename="gemma-4-e2b-it.BF16-mmproj.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use MassivDash/Gemma-4-typescript-coder with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

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 MassivDash/Gemma-4-typescript-coder:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

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 MassivDash/Gemma-4-typescript-coder:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

Use Docker

docker model run hf.co/MassivDash/Gemma-4-typescript-coder:Q4_K_M

LM Studio
Jan
Ollama
How to use MassivDash/Gemma-4-typescript-coder with Ollama:
```
ollama run hf.co/MassivDash/Gemma-4-typescript-coder:Q4_K_M
```

Unsloth Studio

How to use MassivDash/Gemma-4-typescript-coder with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for MassivDash/Gemma-4-typescript-coder to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for MassivDash/Gemma-4-typescript-coder to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for MassivDash/Gemma-4-typescript-coder to start chatting

How to use MassivDash/Gemma-4-typescript-coder with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "MassivDash/Gemma-4-typescript-coder:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use MassivDash/Gemma-4-typescript-coder with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf MassivDash/Gemma-4-typescript-coder:Q4_K_M

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 MassivDash/Gemma-4-typescript-coder:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use MassivDash/Gemma-4-typescript-coder with Docker Model Runner:
```
docker model run hf.co/MassivDash/Gemma-4-typescript-coder:Q4_K_M
```

Lemonade

How to use MassivDash/Gemma-4-typescript-coder with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull MassivDash/Gemma-4-typescript-coder:Q4_K_M

Run and chat with the model

lemonade run user.Gemma-4-typescript-coder-Q4_K_M

List all available models

lemonade list

Gemma-4-TypeScript-Coder : GGUF

This model is a specialized fine-tune of Gemma 4, engineered for TypeScript-centric web development, strict type safety, and modern full-stack architectures. It was trained using Unsloth Studio for maximum efficiency and precision.

🟦 TypeScript Mastery

This fine-tune specializes in:

Strict Type Systems: Expertise in complex generics, utility types, and advanced interfaces.
Modern Frameworks: High proficiency in Next.js, React, Vue 3, and Node.js.
Visual Logic: Leverages vision-language capabilities to transform UI wireframes or screenshots directly into type-safe components.
Best Practices: Focus on clean architecture and idiomatic TypeScript patterns.

🤝 Credits & Acknowledgments

A major shout-out to mhhmm for the typescript-instruct-20k dataset. This robust collection of instructions allowed the model to grasp the nuances of the TypeScript ecosystem effectively.

🚀 Usage & Inference

The model is provided in GGUF format, compatible with llama.cpp.

Example usage:

Standard Text Chat: llama-cli -hf MassivDash/Gemma-4-typescript-coder --jinja
Vision/Image Tasks: llama-mtmd-cli -hf MassivDash/Gemma-4-typescript-coder --jinja

📂 Available Model Files

gemma-4-e2b-it.Q8_0.gguf
gemma-4-e2b-it.BF16-mmproj.gguf

⚠️ Ollama Note for Vision Models

Important: Ollama currently requires a unified blob for vision models.

To use this with Ollama:

Ensure your Modelfile is in the same directory as the merged BF16 model.
Run: ollama create model_name -f ./Modelfile

🔗 Stay Connected

For more insights on AI development and fine-tuning, visit my blog: 👉 spaceout.pl

This model was trained 2x faster with Unsloth

Downloads last month: 1,306

GGUF

Model size

5B params

Architecture

gemma4

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW

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