GPT-4.1 support - GPT-4.1 is now available for all users who have access to model selector dropdown on following agent capabilities: Optimal Choice, Highest Relevance, Complex Reasoning. On average, GPT-4.1 is capable of providing a bit faster and more conversational responses than GPT-4o.
GPT-o4 mini support - GPT-o4 mini has replaced the previous GPT-o3 mini model for Complex Reasoning agent capability, in all three modes: Fast, Balanced and Smart. On top of increased reasoning capabilities, this comes with the additional benefit of supporting numbered citations when selected.
Unlimited conversation time window - Premium users can now allow agents to access conversation history older than 24 hours, enabling chats of potentially unlimited duration. Agents are still bound by their history token window.
🛠 Improvements
Updated styling for all in-chat error messages.
Copy on the new Starter Questions card is now automatically adjusted for the agent's selected language.
Agents newer than 24 hours now have Today as the selected range on Analyze and Dashboard pages.
Added two new states for agent processing feedback: Account limit reached and Success state variant with not all pages successfully processed.
Improved UX of agent processing feedback widget and model when on Ask page.
Improved general flow and page redirects for agent processing feedback.
🐞 Bug Fixes
Fixed the issue where the View more button on the Starter Questions card was not visible on white background.
Fixed the issue where Customer Intelligence custom range selector was always offset by one day if the user's timezone is behind UTC.
Fixed the positioning issue of Developers button on mobile screens.
Fixed the issue with the Microsoft Account button missing on one of the login screens.
Fixed the issue where agent title color was sometimes missing.
Fixed the issue with numbered references format on some models.
Fixed the issue with the free sitemap finder tool that was causing it to fail.
Customer Intelligence data export - All data available in the Customer Intelligence feature can now be exported in full detail.
New Starter Questions layout - Starter Questions layout has been updated to increase visibility and user engagement, and to make agents with a large number of questions, or with particularly long questions, more enjoyable to use.
Automatically generated Personas - All new agents created using sitemap or URL will now have their Personas automatically generated.
🛠 Improvements
Introduced a number of performance improvements when exporting a large amount of data, including breaking data into chunks and adding them in a single ZIP file.
Revamped styling of Share and Export Conversation modals.
Improved styling of indexing feedback widget and modal.
It is now possible to access an agent's page by clicking on its name and avatar on the All Agents page.
🔌 API & MCP
A new Developers page has been added to the app, intended for API keys management and to serve as a resource center.
Numbered citations are now supported when making requests via API.
When retrieving messages via API, it is now possible to choose whether to include numbered references (if present).
API keys per account limit has been increased from 5 to 20.
Revamped tool names and descriptions for MCP server, improving general usability and understanding by third party AIs.
MCP access tokens are now automatically generated and assigned to agents.
A number of smaller improvements and tweaks for MCP servers.
🐞 Bug Fixes
Fixed the issue with the agent’s shareable slug not being correctly generated when duplicating the project.
Fixed the issue where agent preview on Personalize page was not being automatically reset when Persona change is made.
Fixed the issue with the agent loader not being visible if the background is white.
URLs for exported files are now shorter and easier on the eye.
Agent indexing feedback will now gradually display found pages when creating the agent using sitemap.
A number of improvements were released for our MCP server including improving the authentication method, improving connection stability and adding configuration for Trae IDE.
A new layout has been applied to all tabs in the app.
🐞 Bug Fixes
Fixed the issue where very large sitemaps were not fully processed when re-syncing.
Fixed the issue with agent prompt field height on mobile devices, when the user prompt is very long.
Fixed the issue with document metadata sometimes not loading properly.
MCP Servers - All CustomGPT.ai agents now come with a hosted MCP server, enabling one-click integration with any AI agent which supports MCP.
Customer Intelligence feed & charts - On agent’s Analyze page, we now have latest prompts and latest prompts the agent couldn’t answer are available as a live feed. Charts with Content Source, User Emotion, User Intent and Query Language have also been added, with selectable ranges.
Microsoft Accounts support - Users can now connect their existing CustomGPT.ai accounts, or create new ones, using their Microsoft Account’s OAuth.
New no-code deployment options - Added in-app support for deployment on Pendo and Notion, as well as extended support for iFrames.
🛠 Improvements
Customer Intelligence filters are now dynamically added to the URL when changed, enabling sharing of the page with team members.
Agent’s settings pages have been reorganized to enable easier and more logical access to features.
In-agent recaptcha notice is now positioned below the user prompt field, and has automatically adaptable font color.
Values passed with queries using CRM Integration/External ID feature are now visible as the end-user's name on agent’s Ask page.
A quick access link to CustomGPT.ai Slack Community is now available in the sidebar and on the Resources page.
Improved Complex Query Breakdown system prompt.
🐞 Bug Fixes
Fixed the issue with agent layout sometimes not respecting the screen and/or iFrame boundaries correctly.
Fixed the issue with citation links sometimes not working when metadata URL was manually changed.
Fixed the issue with Customer Intelligence filters being reset after going to a different Analyze tab.
Exported JSON files will now correctly be downloaded, rather than opened inside the browser.
Fixed styling issues on My Profile button.
Fixed the issue with chat and sidebar height on Ask page.
Fixed the issue with Instant Viewer code snippet.
Fixed the issue with “Send to Dev” buttons not working.
Fixed the issue with left-aligned Live Chat not using the full screen on mobile devices.
Fixed the issue with query languages not always displaying correctly on Customer intelligence.
Fixed the issue with Not used value not being correctly displayed on Customer Intelligence.
Exports are now sent to the email - All exports from the app are now sent to the user's email, enabling secure and stable exporting of any amount of data.
🛠 Improvements
Custom Agent Title on the Agent Appearance page now has customizable color.
Reviewed and improved the quality of all Customer Intelligence metrics.
Change Content Source Found metric name to Content Source, Yes value to Found and No value to Not Found in Customer Intelligence.
Added new Not Used value to Content Source, and Unclear value to Emotion and Intent metric in Customer Intelligence.
The top bar has been removed from the entire app, to provide more working space for the users.
Simplified no-code instructions for “Deploy with CustomGPT.ai button” option for Website Copilot and Instant Viewer deployments.
If features are enabled, Share and Export icons are now visible in Agent Settings Preview, regardless of whether there is active conversation or not.
OpenAI SDK queries are now correctly displayed in the “Sent From” chart on the Analytics page.
Auto-generated Starter Questions will now always respect the language of source material.
Improved copy for the feature “Should the agent mention source names?” to improve clarity.
Increased maximum available time for processing a single sitemap.
🐞 Bug Fixes
Fixed the issue where Integrations and Advanced deployments pages were visible even for private agents, displaying deployments which are not available on private agents.
Fixed the issue with sitemaps sometimes using too much memory and ending up stuck in the syncing process.
Fixed the problem where a single page without sitemap was sometimes not correctly crawled.
Fixed the issue with Confluence pages not always being correctly auto-synced.
Fixed the issue with SCIM endpoints for SSO login.
Fixed the issue with Affiliate ID not being correctly used in agent UI.
Fixed the issue with documents sometimes not being defined while providing processing feedback.
Fixed the issue with files sometimes not being correctly saved.
Feedback while creating new agents - Added tracking for document processing while creating new agents - users can now see how many documents (web pages, files, etc.) have been found for provided source, and how many have been fully processed so far.
🐞 Bug Fixes
Fixed the issue with the sitemap crawling system not being able to handle nested PHP sitemaps.
Fixed the issue with Customer Intelligence feature when results have more than 10,000 rows.
Fixed the issue where utm_content tag was never added on links provided in agent’s messages.
Fix the issue with SCIM login in cases when certain Identity Providers were not sending the correct value for userName.
Fixed the issue where users could continue messaging the conversation that was shared to them.
Fixed the issue when account limits are hit while connecting Zendesk integration.
Fixed the issue where the “Allow agent to mention sources” setting was not always respected.
Fixed the issue with broken elements on Dashboard after subscription has expired.
Fixed the minor color issue for in-page tabs on mobile devices.
OpenAI API compatibility mode - Introduced new API endpoint, app.customgpt.ai/api/v1/projects/{agent_id}/chat/completions, compatible with OpenAI endpoint api.openai.com/v1/chat/completions, improving accessibility of our API ecosystem.
🛠 Improvements
Support for GPT-4-Turbo has been removed, and all agents which utilized that model have been switched to GPT-4o.
Agent and user avatars can now be aligned to the left or right edge of the screen independently of each other.
Starter questions styling has been updated.
The in-chat user avatar is now using colors consistent with the agent's color scheme.
The in-chat scrollbar has been moved to the edge of the screen.
We are now adding utm_content instead of utm_source tag to outgoing links in chats.
Improved Agent Settings page navigation on mobile devices.
🐞 Bug Fixes
Fixed problem with agent settings not always being reflected in real time.
Fixed the issue with KaTeX formulas not always being displayed correctly in chats.
Fixed the issue with RAR archives appearing in cited sources.
New Share Link format - All agents now have a Share Link deployment URL in the format agentslug.customgpt-agents.com, which is the same for both private and public agents.
UTM tags on in-chat links - We are now adding utm_source=customgpt.ai to all citations, and to links directly provided by agents in chats.
AB testing support - We now support running basic AB tests on users, by assigning them to different groups and changing conditions in the app based on their group.
OpenAI fallback - In case of OpenAI API service being down, our agents will now automatically switch to Azure or Anthropic providers, ensuring service for end-users.
🛠 Improvements
Links to various resources have been moved from the dashboard to the separate Resources page, along with additional support and sales related resources.
Improved handling for KaTeX formulas inside chats.
Extended the recovery URL functionality for abandoned Stripe carts.
Smaller backend improvements.
🐞 Bug Fixes
Fixed the issue with the send button being part of the scrollable content in the agents’ prompt field.
Fixed the issue with in-agent Terms of Service not being displayed if white-labeling is activated.
Fixed the issue with citations when sharing the conversation.
Serving all assets from Amazon S3 - We are now serving all app’s assets (images, scripts, etc) from Amazon’s CDN (Content Delivery Network), enabling consistent and scalable content delivery across the globe.
Finalized o3-mini support for Complex Queries capability - o3-mini (balanced mode) is now the default AI model when Complex Queries capability is used, and is available for all Premium users.
OAuth Support - we now support users providing authorization via OAuth, rather than using API keys. This is currently a hidden feature which will be expanded into the proper product over time.
Deepseek Support - Deepseek R1 model is now supported in our dev environment.
🛠 Improvements
Porn sites can no longer pass the sitemap check and be added for crawling.
Ask Me Anything now loads conversations instantly even for huge projects.
When Complex Query Capability is being used, agents can now dynamically decide should they break down the query or not.
Captcha notice is no longer visible inside agents if conversation has started.
Improved error messaging when a user attempts to sync Google Drive files larger than 500MB.
The Customer Intelligence tab is now capable of receiving filter parameters from the URL.
The Export format on the Analytics/Breakdown chart has now been standardized, regardless of which sub-chart is being selected.
It is now possible to update project’s Share and Export settings via Zapier integration.
Agent UI and UX improvements - Updated styling and appearance of all deployed agents, with the aim of improving end-user experience. List of major changes:
All deployments now have consistent design, and user experience is standardized across deployments.
Mobile experience is greatly improved.
Added font breaking points, where smaller agents now use reduced font size to display more information.
Breakpoints now depend on agent size rather than screen size.
User messages now don't use full-width anymore, and are right-oriented.
Added the option to use background color instead of the image.
On large agents we now display up to three starter questions on the screen.
Starter questions UI has been improved.
Initial agent state is now styled differently depending on if there is a starter question(s) or not.
Inline citations styling has been updated and improved.
A number of smaller UI improvements.
🛠 Improvements
Starter questions are now automatically generated for File Upload project source.
Improved SEO tags, descriptions and icons for Login and Register pages.
Added frontend latency tracking, to help improve end-user experience.
🐞 Bug Fixes
Pagination will now correctly reset when changing filters on the Data / Documents page.
Resolved issue with Customer Intelligence data not updating in real time.
Fixed issue where inline citation tags were included in export from Ask Me Anything page.
Fixed issue with the wrong base URL for cited Confluence pages.
Fixed issue where on auto-sync, Notion pages were being deleted before new ones were added.
Fixed remaining issue with overlapping ticks on Breakdown charts.
Fixed error handling with deleted files in Google Drive.
Fixed remaining UI issues related with SharePoint disconnected alert.
Fixed issue with starter questions generator and big files.
