Instructions to use Roxas13/gemma3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Roxas13/gemma3 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("Roxas13/gemma3") 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
- MLX LM
How to use Roxas13/gemma3 with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "Roxas13/gemma3"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "Roxas13/gemma3" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Roxas13/gemma3", "messages": [ {"role": "user", "content": "Hello"} ] }'
| license: gemma | |
| library_name: mlx | |
| pipeline_tag: text-generation | |
| base_model: mlx-community/gemma-3-4b-it-4bit | |
| base_model_relation: quantized | |
| tags: | |
| - mlx | |
| - gemma3 | |
| # AEGIS Gemma 3 4B IT (text-only, 4-bit MLX) | |
| Text-only, 4-bit MLX-quantized build of Gemma 3 4B IT, used on-device by the | |
| AEGIS app. This is a **modified** (vision tower removed, 4-bit quantized) | |
| derivative of Google's Gemma 3 4B IT. | |
| ## License & notices | |
| Gemma is provided under and subject to the **Gemma Terms of Use**: | |
| https://ai.google.dev/gemma/terms | |
| Use of this model must comply with the Gemma **Prohibited Use Policy**: | |
| https://ai.google.dev/gemma/prohibited_use_policy | |
| This repository redistributes a modified Gemma derivative. By Google's Gemma | |
| Terms, the above Terms of Use and Prohibited Use Policy apply to this model and | |
| are passed along to all recipients. Modifications relative to the base model: | |
| vision encoder removed (text-only) and weights quantized to 4-bit for MLX. | |