Frequently Asked Questions
Quick answers to common questions about the Iris platform
General Questions
What is Iris?
Iris is an autonomous AI agent platform built by irisvision.ai. Unlike traditional chatbots that only answer questions, Iris takes real actions on your behalf -- researching topics, generating documents, building presentations, creating websites, analyzing images, and much more. You describe a goal in natural language and Iris plans, executes, and iterates until the task is complete.
How does Iris differ from ChatGPT?
ChatGPT is primarily a conversational interface for a single LLM. Iris is an agent platform that orchestrates multiple AI models and over 20 integrated tools to complete multi-step tasks autonomously. Where ChatGPT answers questions, Iris takes action -- producing tangible deliverables like reports, slide decks, and deployed websites rather than just text responses.
What models does Iris use?
Iris uses a multi-model architecture powered by LiteLLM. It intelligently routes tasks to the best available model, including Anthropic's Claude for nuanced reasoning, OpenAI's GPT-4 for creative generation, and specialized vision models like GPT-4V and Claude 3 for image analysis. The platform selects the optimal model for each step of your workflow automatically.
Is Iris free?
Iris offers a free tier with a monthly credit allowance so you can explore the platform at no cost. For heavier usage, Pro and Enterprise plans provide higher limits, priority processing, and dedicated support. All plans use usage-based billing so you only pay for what you consume.
Capabilities
What can Iris research?
Iris can perform deep web research on virtually any topic. It searches the web, browses pages, extracts structured data, and synthesizes findings into comprehensive reports with proper citations. Common use cases include competitive analysis, market research, academic literature reviews, people research, and trend monitoring.
Can Iris create presentations?
Yes. Iris can auto-generate slide decks from any topic, data set, or document. It structures content into slides, applies professional formatting, and exports to PPTX format. You can specify the number of slides, tone, target audience, and visual style. The presentation service handles layout, bullet points, charts, and embedded images automatically.
Can Iris build websites?
Yes. Iris can generate and deploy web pages through natural language. Describe the page you want -- a landing page, portfolio site, dashboard, or form -- and Iris will produce clean HTML, CSS, and JavaScript. It can deploy the result directly so you get a live URL to share immediately.
What image formats does Iris support?
Iris supports PNG, JPEG, GIF, and WebP for both input and output. You can upload images in any of these formats for analysis using vision models (GPT-4V and Claude 3), and generated images are delivered in PNG or JPEG. The vision pipeline also handles screenshots, charts, diagrams, and scanned documents.
Account & Billing
How do I create an account?
Visit irisvision.ai and click Sign Up. You can register with your email address or use a social login provider. Once signed up, you will have immediate access to the platform and can start using Iris right away with the free-tier credit allowance.
What payment methods are accepted?
Iris processes payments through Razorpay, which supports credit cards, debit cards, UPI, net banking, and popular digital wallets. All transactions are handled securely through Razorpay's PCI-DSS compliant infrastructure. You can manage your payment method from the billing section of your dashboard at any time.
Can I cancel anytime?
Yes. There are no long-term contracts. You can downgrade or cancel your plan at any time from your account settings. If you cancel a paid plan, you will retain access to your current tier until the end of the billing period, after which your account reverts to the free tier. Your data is preserved and accessible regardless of plan status.
How is usage tracked?
Iris tracks usage in real time across three dimensions: LLM tokens consumed, vision API calls, and tool invocations. You can monitor your current usage and remaining credits from the billing dashboard. Detailed breakdowns show per-task costs so you always know exactly what you are spending.
Technical
Is my data secure?
Yes. Iris stores all data in Supabase (managed PostgreSQL) with Row-Level Security (RLS) policies that ensure users can only access their own data. All API communication is encrypted over HTTPS, and API keys are hashed before storage. The platform supports team-based access controls through Basejump multi-tenancy, and you can delete your data at any time from the dashboard.
What file formats can Iris process?
Iris can process a wide range of input and output formats. Input: PDF, DOCX, PPTX, CSV, XLSX, TXT, PNG, JPEG, GIF, and WebP. Output: PDF documents, DOCX files, PowerPoint presentations (PPTX), CSV spreadsheets, HTML web pages, and images in PNG and JPEG. The document processing pipeline supports rich formatting, tables, charts, and embedded images.
Does Iris have an API?
Yes. Iris provides a RESTful API and an official Python SDK for programmatic access. You can create threads, send messages, retrieve responses, manage conversations, and trigger agent tasks from any application. Authentication uses API keys that you can generate from your dashboard. See the API Reference and SDK pages in this documentation for full details and code examples.
Can I use Iris offline?
No. Iris requires an active internet connection to function. The platform relies on cloud-hosted LLM providers (Anthropic, OpenAI), web browsing tools, and server-side processing for document generation and tool execution. However, all generated outputs (documents, presentations, images) can be downloaded for offline use once they are created.
Troubleshooting
Why is my task taking long?
Complex tasks that involve multiple steps -- such as deep web research with many sources, large document generation, or multi-page website builds -- naturally take longer to complete. Iris executes these tasks autonomously in the background, so you can navigate away and return when the task is finished. If a task seems stuck for more than a few minutes, try refreshing the page or starting a new thread. High server load during peak hours can also contribute to longer processing times.
Why did my generation fail?
Generation failures can occur for several reasons: rate limits from upstream LLM providers, insufficient credits in your account, overly complex requests that exceed token limits, or temporary service interruptions. Check your billing dashboard to verify you have available credits, then try simplifying your request or breaking it into smaller steps. If the problem persists, contact support with the thread ID so we can investigate.
How do I contact support?
You can reach the Iris support team by emailing support@irisvision.ai or by using the help widget in the bottom-right corner of the dashboard. For faster resolution, include your account email, the thread ID of the affected conversation, and a description of the issue. Enterprise customers have access to priority support with guaranteed response times.