Skip to main content
Welcome to the official changelog for Irona AI. Here you’ll find a history of product updates, improvements, and fixes. We’re continuously evolving — and we want you to stay in the loop.
We’re deeply proud of version 1.0, the culmination of long hours, smart engineering, and one million data‑driven lessons. This is more than a feature launch, it’s the start of an intelligent platform built for you.

​
Changelog • Version 1.0 • 2025‑07‑30

What’s new in our first release Added: Unified chat interface for top AI & LLM models
Access all popular LLMs in one interface, no need to juggle platforms!
Automatic LLM Routing chooses the best model per query based on performance, cost, or latency. Everything runs dynamically and seamlessly. Added: Hyper‑personalization via user feedback
You can now thumbs‑up or thumbs‑down individual responses.
Our router learns from your feedback and tailors future routing to match your preferences, so it gets smarter the more you use it. Added: Routing optimized from 1M+ training data points
Our backend intelligently routes your requests across LLMs using optimized logic, trained on over one million data points, to deliver top performance consistently. Added: Arcade Mode for comparing AI model outputs
Run multiple models in parallel and explore side‑by‑side responses.
Pick your favorite response, no guessing, just interactive model comparison.

​
In short, we’ve got you covered from the very start

One start, infinite reach. One interface, all your favorite LLMs in one place.
Personalized AI, that gets your style through feedback.
Arcaded comparison for curated model selection.
A solid foundation built from real user data. We’re just getting started.

​
And, there’s more on the way

We’re building even bigger tools to empower you: 🎯 Search: retrieve web or internal content directly. 📄 PDF support: feed documents and get smarter AI understanding. 🔬 Deep Research tools: structured data, citations, analysis. 🤖 MCP & Agents: multi‑chain prompting and autonomous agents. 🧠 Routing across all of these: choose the best path for every use case and query. Welcome to the future of chat interfaces, centered on choice, performance, and you.
Thank you for being part of our journey. Let’s keep building greatness together.

​
2025‑08‑06

Added: Support for Anonymous/Guest Users
Start exploring Irona AI instantly without sign-up friction. Ideal for quick experiments or first-time users evaluating the platform. Launched: Irona Labs AI Studio
A powerful playground for developers, researchers, and AI tinkerers. Customize prompts, test model responses, view token costs, and monitor latency in real time, all in one advanced dashboard.

​
2025‑08‑13

Added: New Models Support
Added support for the latest AI models including GPT-5 series, GLM-4.5 series, and Kimi K2. Expand your AI capabilities with cutting-edge language models. Improved: Bug Fixes & Latency Optimizations
Dramatically improved website and the chat interface speed through comprehensive bug fixes and latency optimizations. Experience faster response times and smoother interactions. Improved: Mobile Responsiveness & UI Consistency
  • Enhanced mobile responsiveness for ChatsBar, Chat, Dashboard, and ModelSelector components
  • Added mobile overlay backdrop, improved sidebar toggle behavior, refined padding and font sizes
  • Introduced mobile header with menu dropdown for easier navigation
  • Improved mobile menu toggle and dropdown functionality across all components

​
2025-08-20

Added: Throughput Tracking
Introduced throughput monitoring to give better insights into system performance and request handling. Added: Rebranded Domains
ironlabs.ai: Our main landing page.
iron.cx: Our chat-app.
Both are now live and in healthy condition. Improved: Background Job Handling
Enhanced scheduling with new cron job functionality to improve automation and background task reliability. Fixed: Anonymous Feedback Flow
Resolved issues in feedback submission for anonymous/guest users, ensuring smoother feedback collection. Fixed: Emoji Rendering
Improved emoji rendering for onboarding and general usage, ensuring consistency across the platform.

​
2025-09-03

Added: Seamless Dodo Payment Integration
Enabled smooth payment setup for both custom and auto top-ups, enhancing convenience for end users. Added: Keyboard Shortcuts
Introduced a full set of keyboard shortcuts to streamline navigation and improve productivity in the chat app. Improved: Model Selector Modal
Redesigned the Model Selector UI for a more intuitive and user-friendly experience. Improved: Cold-Start Latency
Implemented a Keep-Alive server (via Vercel) to reduce cold-start delays and deliver faster response times.

​
2025-09-17

Added: Regenerate Feature
New regenerate functionality allows users to regenerate responses from a different or same model if the user doesn’t like a response, even in the middle of a conversation thread. Thereby providing the users an enhanced conversation flow control. Added: UX Enhancements
When regenerating or streaming responses, the chat window no longer auto-scrolls. Instead, a “Scroll to Bottom” button appears, giving users enough time to read through the response at their own pace, delivering a smoother, more controlled experience. Fixed: UI Issues
Resolved various UI inconsistencies and bugs, ensuring a cleaner and more reliable interface.

​
2025-09-24

Added: Reasoning Token Support
Introduced full support for reasoning models with live reasoning & reasoning summary.
Price and usage calculations now account for reasoning tokens, ensuring accurate billing and reporting.

​
2025-10-01

Fixed: Irona Studio UI Issues
Addressed UI bugs and inconsistencies in Irona Studio for smoother workflows and better user experience. Enhanced the profile section for clarity and stability. Fixed: Subscription Validation
Resolved subscription validation errors to ensure reliable access and accurate billing. Stabilized chat subscription logic to prevent failures and ensure consistent usage tracking.

​
2025-10-08

Added: Image Generation Mode
You can now generate high-quality images using prompts, seamlessly integrated into the chat interface. Perfect for prototyping, design ideation, and creative tasks. Added: New Reasoning Models Support (Chat App & SDK)
Enhanced outputs with reasoning capabilities from GPT-5 & newer Grok models, improving trust, reliability, and actionable insights for end users.
⌘I