Shadow
Quill
Your Private...
Prompt Engineer
Use powerful open weight LLMs to upgrade your prompts and ideas today. Transform your basic prompts into perfectly structured AI chat/agent inputs. Open weight models like Gemma 3 are free for public download. ShadowQuill also has no cost and will remain free forever.
100% Private & Offline
Your prompts and data never leave your computer. All processing happens locally, giving you complete control and confidentiality.
Powered by Gemma 3
Gemma 3 is Google's open-weight model. This means it runs completely through your computer's hardware, not the cloud. Check out Google's resources on it here.
Simple & Accessible
Runs on any modern computer with at least 8 to 12 GB of RAM. Connects to your local Ollama Gemma 3 instance in seconds.
Your Data Stays With You
Absolute Privacy. No Cloud.
ShadowQuill respects your complete privacy. This is our core principle. There are no servers. There is no cloud processing. We collect no data. We use Gemma 3 within Ollama. Your prompts and your ideas stay on your computer.
Own the Technology
Most AI tools require expensive subscriptions. Our goal is to offer you a better path. You will never pay a monthly fee for AI usage in ShadowQuill. You gain complete ownership of the AI technology. It runs on your own computer's hardware. It is 100% free forever.
Reccommended Minimum Requirements
Choose Your Local AI Power
Select the right Gemma 3 model for your machine. Learn what your hardware needs to run ShadowQuill completely offline.
Supported Gemma 3 Models
4B Parameter Model Best balance
RAM Required: 8 to 12 GB or more
This model is fast. It uses fewer resources. It runs well without a dedicated GPU.
12B Parameter Model
RAM Required: 16 GB or more
Choose this model for complex work. Stable use requires a modern GPU or Apple Silicon.
27B Parameter Model
RAM Required: 32 GB or more
This is our most powerful option. A high-end GPU like an RTX card is strongly recommended.
About Gemma 3
- Google built Gemma 3 as a free to use open weight model and trained it with the same technology used for the powerful Google Gemini models.
- The models can process text and images. They are multimodal.
- The models support many languages. This includes over 100 languages.
- It works as a private offline chat via Ollama or LLM Studio. It also fine tunes prompts within ShadowQuill.
Your Two-Step Launch Plan
OL
Install Ollama
Ollama is the local AI engine. It is required to run Google's Gemma 3. Download the version for your system now. Then install the model.
Start Ollama. Then run these commands to download a supported Gemma 3 model:
ollama pull gemma3:4bollama pull gemma3:12bollama pull gemma3:27b
Install ShadowQuill
ShadowQuill installs easily and automatically finds your local Gemma 3 models in Ollama. Select your platform below.
Proudly Open Source
ShadowQuill is a passion project from AI enthusiasts, built for users who believe in local-first privacy.
View on GitHub
Reach Out to Us
Have a question, suggestion, or a project in mind?
Our inbox is always open. Let's connect!