Gemma 3 (270M) run on Ollama using a quad-core CPU
Free platform to access multiple AI models
Pocket TTS optimized for Hugging Face Spaces on CPU