I have often that GPT offers to me to create and attach the document. That is then done, it says, but the document is nowhere to find. In the ChatGPT app itself, this works perfectly and documents are attached within the chat and can be clicked and opened/downloaded.

At the moment, there is a limit of 30 memories for multi-chat. Would be great to increase the level to 60 or 90 for Business and Entreprise plan.

- CURRENT MEMORY IN CHAT IS TOTALLY IMPRACTICAL 👎

: Although there is a tag system to organize and search information, What is the point of having conversations and assets sorted in tags if we cannot start a new chat reusing tag-sorted data ?

💡 Please add the possibility to start a new chat reusing tag-sorted data (chats and files). This will allow to easily start chatting with predefined memory information and related files, as Context/ Knowledge Base 🧠, in order to improve the chat results and user experience

⁀➴ : Start new chat from an existing selected tag → the corresponding are automatically configured as and as

Also, it will help to organize and sort chats inside folders 📁 directly on the conversation history sidebar and be able to start a new chat from a folder as Knowledge Base (aka Project), like in this video https://www.youtube.com/watch?v=Pmb3y3GPbd8&t=819s

In general, 1min AI still lacks an effective and easy-to-use Knowledge Base for new chats.

Gemini 2.5 Flash is a model that can be used for free (with fair usage) in different AI platforms. With all respect, quite frankly I don't understand the astronomical high credit cost of this model in 1min AI.

For the same prompt and output size: Gemini 2.5 Flash uses double the amount of credits than GPT-5 mini and ❗️ more credits than GPT-5 nano

Could you please optimize the token cost of this model ? To match cost of performance equivalent models like GPT-5 nano/mini

add the ability to one click delete memories, click on trash one by one is boring.

Adding the new MCP to the platform.

As described by Anthropic:

MCP is an open protocol that standardizes how applications provide context to LLMs. Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools.

hi can you add the deep search option to the multichat ai with ai engines ?