ThreoAI is Synthreo's secure, organization-managed AI chat platform. You chat with leading AI models, build reusable Experts, run Builder Agents, and organize work in Projects, all within an environment your administrator controls.

What is an Expert in ThreoAI?

An Expert is a reusable AI assistant set up once with instructions, optional training documents, and actions. You build Experts in Studio and drop them into any chat. Experts were previously called Custom GPTs.

What file types can I upload to ThreoAI?

ThreoAI supports PDF, DOCX, TXT, PPTX, images, XLSX, and CSV uploads. The AI can read, summarize, extract data from, and answer questions about uploaded content.

How does ThreoAI connect to Synthreo Builder?

Builder powers two things in ThreoAI. Builder workflows back the actions you wire onto an Expert, and Builder agents are surfaced in chat as Agents that run multi-step tasks for you.

What is Model Comparison in ThreoAI?

Model Comparison sends one prompt to two AI models at once and shows both answers side by side, so you can see which model handles a task best.

ThreoAI Overview

ThreoAI platform overview - chat with many models, build Experts, run Agents, create in Studio, add Connectors, and organize work in Projects, all governed.

Overview

ThreoAI is Synthreo’s secure, organization-managed AI platform. You chat with leading AI models, build reusable Experts, run Agents, and keep work organized in Projects - all inside an environment your administrator controls.

The difference from a consumer AI tool is governance. Your administrator decides which models are available, who has access, and how the platform is set up. You get the same conversational experience, and your IT and compliance teams keep their oversight.

ThreoAI main interface

Platform capabilities

Chat with many models

Pick a model from the selector at the top of the composer and start typing. Each model carries capability badges that show what it can do - web search, image generation, charts, and more. You can switch models in the middle of a conversation when another one fits the task better.

Use the chat for everyday work: drafting emails and reports, answering questions, brainstorming, generating code, translating, or summarizing what you paste in. Quick-action chips above the composer (Summarize, Analyze, Draft, Brainstorm, Explain) give you a fast start. Every conversation saves to your sidebar history.

See Starting a Chat and Tools and Capabilities.

Experts

An Expert is a reusable assistant tuned to one job. It chats like a normal conversation but follows the instructions and knowledge its creator gave it. You can build your own or use ones shared with you.

When you build an Expert you set:

instructions that define its role, tone, and rules,
optional training documents it treats as a knowledge base,
actions it can take, backed by Builder workflows, and
who can use it.

Experts were previously called Custom GPTs. See Creating an Expert and Using an Expert.

Agents

Agents are Builder agents surfaced in chat. Where an Expert shapes how a model answers, an Agent runs a plan - it uses tools and works through multi-step tasks for you. Your administrator shares Agents with your organization, and they show up in the Agents catalog and the Marketplace.

Agents are documented in the Builder section. See Welcome to Builder.

Studio

Studio is the creation hub. Once you build something there, it shows up across chat, Projects, and the Explore catalog. Studio has tabs for Overview, Experts, Agents, Connectors, and Projects, and it is where you create Experts, Connectors, and Projects.

See Studio Overview.

Connectors

Connectors bring external systems into your chat through MCP - search engines, calendars, ticketing tools, internal APIs. Pick from a catalog or point at any MCP server URL to add your own. Once connected, the AI can act on those systems from inside a conversation.

See Connectors.

Projects

A Project groups related conversations and shared context together, so a thread of work keeps memory across chats. Use one per client, topic, or workstream. Conversations inside a Project stay there instead of cluttering your general history.

See ThreoAI Projects.

Library

The Library holds web apps and background jobs you can run from chat. It is available where enabled, so you may or may not see it depending on your organization’s configuration.

See Library Overview.

Model comparison

Model Comparison sends one prompt to two models at the same time and lays their answers side by side. It is the quick way to see which model summarizes a document better, writes cleaner code, or handles a tricky question. Follow-up messages go to both models.

See AI Model Comparison.

Files

Attach files to any conversation - the home chat, a Project, or an Expert.

Supported formats: PDF, DOCX, TXT, PPTX, images (PNG, JPG, JPEG, GIF, WEBP), XLSX, CSV.
The AI reads, summarizes, extracts data, and answers questions about what you attach.
Attach several files at once to compare or cross-reference them.

See Working with Files.

Memory and personalization

ThreoAI remembers facts from your chats and applies them later, and you can shape that from your Profile. Set a preferred name, describe your role, and decide what the AI keeps. You can also import memories from ChatGPT, Claude, or Gemini, so you do not start from scratch.

See Memory and Personalization.

Tweaks

Tweaks is the appearance panel in the sidebar footer. It controls theme, accent color, corners, spacing, font, text size, and more - purely visual, applied instantly, and personal to you.

See Tweaks and Appearance.

Common use cases

IT support. An Agent or Expert answers from your runbooks and SOPs so people get help without filing a ticket.
Document analysis. Upload contracts, policies, or reports and ask for a summary, the key risks, or specific data points.
Internal knowledge base. An Expert grounded in your own documents answers policy questions in plain language.
Onboarding. An Expert walks new hires through benefits, equipment, and setup, one question at a time.
Process automation. An Expert’s actions, backed by Builder workflows, file a ticket or update a record straight from chat.
Model evaluation. Use Model Comparison to pick the right model before you commit to it.
Content creation. Draft proposals, emails, and marketing copy, then refine through conversation.
Data analysis. Upload a spreadsheet and ask for trends, outliers, or a summary of the numbers.

Using ThreoAI with Builder

ThreoAI and Builder work together in two ways. The actions you wire onto an Expert call Builder workflows, so an assistant can reach into ticketing platforms, CRMs, and email services. And Agents are Builder agents surfaced right in chat. Build the workflow or agent in Builder, then use it from ThreoAI.

Welcome to Builder - introduction to the no-code AI agent platform.
Builder User Manual - the full reference for building workflows and agents.
Tenant Management Overview - where administrators manage users, customers, AI model configuration, and permissions.

Welcome to ThreoAI - getting access and first steps
Navigating the Interface - the sidebar, composer, and account menu
Studio Overview - where Experts, Connectors, and Projects are built
Profile Settings - personalization, memory, and your default model