Skip to main content
​
Launch week
Noesis is live. This week’s release brings the full AI chat workspace online, along with subscription billing, isolated spaces, and multi-model access.

​
New features

  • Multi-model chat — Switch between models from OpenAI, Anthropic, Google, and xAI inside a single conversation. Standard, Premium, and Ultra-Premium tiers are available depending on your plan.
  • Isolated Spaces — Organize work into dedicated spaces with their own emoji, system prompt, and chat history. Keep client projects, research, and personal work separate.
  • Web Search (RAG) — Toggle real-time web results on any message. Answers include inline citations so you can verify sources.
  • Automated context compaction — Long conversations are summarized automatically to preserve memory across turns while reducing token cost.
  • Multimodal input — Attach and analyze text files, code, PDFs, and images directly in chat.
  • Subscription plans — Two paid tiers are now available:
    • Synthesis ($14.99/mo) — 1,000 credits/day, access to all models.
    • Noesis ($29.99/mo) — 3,000 credits/day, access to all models.
    • Origin (Free) — 20 credits/day, standard models.
  • Stripe Customer Portal — Manage your subscription, update payment methods, and view invoices from inside the app.
  • Weighted credit system — Daily credit allowances are consumed based on each model’s cost, so lightweight models like Gemini Flash go further than premium ones like GPT 5.6 Sol.
  • Command palette — Jump between spaces, sessions, and actions with a keyboard-driven palette.
  • Keyboard shortcuts — A shortcuts modal lists every available hotkey for faster navigation.
  • Sharing — Share a session with a public link so others can view the conversation.
  • Session threads — Branch a conversation into a side thread without losing the main chat context.
  • Roadmap page — See what’s shipped and what’s coming next at /roadmap.
  • Public newsletter signup — Subscribe from the marketing site to get product updates.

​
Updates

  • Dark mode and glassmorphic UI — The interface ships with light and dark themes and adapts to your system preference.
  • Real-time presence — See when collaborators are active in a shared space.
  • Streaming responses — Model replies stream token-by-token for faster time-to-first-word.