Hey Product Hunt! 👋 I'm thrilled to introduce Fragaria, a project born from the desire to push the boundaries of AI reasoning. Fragaria isn't just another language model wrapper. It's a sophisticated reasoning engine that grows smarter with every problem it solves. Here's what makes it special: 🧠 Chain of Thought (CoT) Reasoning: Fragaria breaks down complex problems into step-by-step solutions, providing transparency in its decision-making process. 🔄 Reinforcement Learning: Our novel approach combines CoT with RL, allowing Fragaria to continuously refine its problem-solving strategies. 🌐 Multi-Provider Support: Easily switch between OpenAI, Groq, and Together.ai as your LLM provider. 📈 Adaptive Learning: A SQLite-based scoring system helps Fragaria remember and improve upon successful strategies. 🔌 Easy Integration: With an OpenAI-compatible API, you can drop Fragaria into existing projects with minimal changes. 🔬 Research-Friendly: Designed for easy experimentation with different CoT strategies and RL algorithms. 🖥️ Interactive Demo: Includes a Svelte-based interface for visualizing the reasoning process. Whether you're a developer looking to enhance your AI capabilities, a researcher exploring new frontiers in machine reasoning, or just an AI enthusiast, Fragaria offers a playground for innovation. Try it out, contribute, or just explore the code: https://github.com/terraprompt/f... I'm excited to hear your thoughts and see what amazing applications you'll build with Fragaria. How would you use an AI that learns to reason better over time?

Fragaria : CoT + RL = Reasoning

Self-improving AI reasoning engine for developers

Self-improving AI reasoning engine for developers