NEURON // LLM Platform

Full-stack multi-model AI platform with auth, conversation history, agent pipelines, model comparison, and Stripe-powered subscriptions. Built with Next.js 16, Supabase, and Pollinations.ai.

Short Explanation

NEURON is a full-stack multi-model AI platform that lets you query several LLMs side-by-side, save conversation history, and run multi-agent pipelines for blog and book drafting. It's built on Next.js 16 (App Router) with Supabase for auth and Postgres, Stripe for subscriptions, and Pollinations.ai as the model gateway - no Anthropic or OpenAI keys required. The UI ships with a neon/cyberpunk theme, dark/light toggle, and is deployed on Netlify.

Some of the features that have been implemented include:

🧠 Multi-model compare: fire one prompt at several LLMs simultaneously on / and read the answers side by side
💬 Chat with history: /chat streams responses, persists to Supabase, supports shareable links and per-conversation routes (/chat/[slug])
🤖 Agent pipelines: /pipeline chains a Writer and Reviewer agent for blog drafts (with editor that flags AI-sounding writing) and a multi-agent book pipeline
💾 Conversations: /conversations lists saved chats with search, delete, and re-open
🔐 Auth: Supabase email/password sign up, login, callback handling, and server-side session via cookies
💳 Subscriptions: /pricing with Stripe checkout for NEURON Pro ( $9/mo) and NEURON Team ($ 29/mo); webhook updates subscriptions table
🖼️ Image + audio generation: /api/image and /api/synthesize (TTS) routes backed by Pollinations
📊 Model cache: /api/models refreshes the available text/image model list every 24h and serves it to the client
🔗 Share links: /share/[id] for public conversation snapshots
🌗 Theming: dark/light/system theme switcher with CSS variables
📱 Responsive layout: keyboard-friendly, mobile-collapsing sidebar on compare and pipeline pages