MemFree Changelog
MemFree is committed to becoming the ultimate AI assistant for Indie Makers and Full-Stack Developers. Our vision is to boost product development efficiency by 10x through cutting-edge AI technology.
Currently, MemFree offers five core features: AI Search, AI Chatbot, AI UI generator, AI Image generator, and AI Coding Assistant.
Moving forward, we will continue to enhance our AI Search and AI Coding capabilities, providing developers with smarter and more efficient solutions to seamlessly transform ideas into products.
This ChangeLog Page is generated by PageGen
Support OpenAI O3-Mini and OpenAI O1 AI Model
Now, MemFree supports OpenAI-O3-Mini and OpenAI-O1 AI Model.
- •OpenAI O3-Mini is more powerful than OpenAI O1-Mini
- •O3-Mini is OpenAI most recent small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini.
- •O3-Mini excels in science, math, and coding tasks.
- •More details about OpenAI O3-Mini can be found here
- •OpenAI O1 is the official version of OpenAI O1-Preview
Support DeepSeek R1 Reasoning Model
The DeepSeek R1 model has the ability to think and reason, is good at solving complex problems, and is good at math and coding tasks.
- •DeepSeek-R1 is DeepSeek's most powerful and newest AI Model.
- •Performance on par with OpenAI-o1
- •Math, code, and reasoning tasks on par with OpenAI-o1
- •More details about DeepSeek-R1 can be found here
AI-Powered Search History
AI-powered search available across titles, questions, and answers
Click the search button at the top of the sidebar
Build the index of historical messages (only once)
Enter keywords or sentences for AI search
Support AI Image Generator
MemFree now supports image generation via the powerful recraft-v3 AI
Input the image description or keywords
Customize the image style, color, and size and use case, which are optional
Click the generate image button
Support DeepSeek-V3 and Google Gemimi-2.0 AI Model
DeepSeek-V3 is DeepSeek's most powerful AI Model, and Gemimi-2.0 is Google's most powerful AI Model.
- •DeepSeek-V3 is DeepSeek's most powerful AI Model.
- •DeepSeek-V3 has extremely fast text generation speed: 60 tokens/second
- •DeepSeek-V3 is available to all users
- •DeepSeek-V3 is more powerful than GPT-4o-mini
- •More details about DeepSeek-V3 can be found here
- •Gemimi-2.0 is Google's most powerful AI Model.
- •Gemini 2.0 has a significantly improved time to first token (TTFT) over 1.5 Flash
- •Gemini 2.0 has better performance across most benchmarks than Gemini 1.5 Pro
- •Gemini 2.0 is available to pro users
- •More details about Gemini 2.0 can be found here
More customization, more optimization
- •Support for Claude 3.5 Haiku AI Model
- •Chat Mode Enhancement
- •Option to control the display of the Mindmap.
- •Ability to delete individual search messages.
- •Automatic generation of search message titles by AI.
- •Display message count and date-time for historical searches.
More customization, more optimization
- •Support for Claude 3.5 Haiku AI Model
- •Chat Mode Enhancement
- •Option to control the display of the Mindmap.
- •Ability to delete individual search messages.
- •Automatic generation of search message titles by AI.
- •Display message count and date-time for historical searches.
Supports OpenAI O1-mini and O1-preview AI Model
OpenAI’s o1 (Strawberry) models are now live on MemFree, These new models are designed to reason first, not just respond, making them ideal for complex queries, especially math and coding. We’re giving early access to our Pro and Premium users.
Support multi languages and multi images
MemFree currently support English, Chinese, German, French, Spanish, Japanese and Arabic. If you would like to see support for other languages, please send us your feedback on the feedback page
- •MemFree will automatically select the language based on your location and browser language settings.
- •You can also manually switch the language in the sidebar on the left.
Increase the maximum size of single file and image that can be uploaded
Increase the maximum size of single file and image that can be uploaded
- •For Free user: Up to 4MB For Single File and Image to Ask
- •For Pro user: Up to 20MB For Single File and Image to Ask
- •For Premium user: Up to 50MB For Single File and Image to Ask
Support search by Image
now, you could search by image in MemFree, you could extract text from the image,you could ask the image content.Both local images and network images are supported!
- •Support ask and summarize Web Pages and PDFs
- •Support login with email
- •Support Google one tap login
Support indexing and AI search local files
If you need more file format, please let us know. Currently, we mainly extract text information from documents.
- •DOCX
- •PPTX
Support saving search history
MemFree supports saving search history now, You can access your search history on any device
- •MemFree automatically decide whether to search the Internet and personal knowledge base now
- •MemFree supports switching light and dark themes
- •MemFree supports Google AI Model and LLama 3.1 AI Model
Support Search, Chat and Ask mode
Support Search, Chat and Ask mode
- •Quickly Access Relevant Content from Your Personal Knowledge Base.
- •Quickly Obtain Webpage Links.
- •Personal Assistant: Coding, Writing, Brainstorming, Translation, and More.
- •Questions on History, Culture, Science, and General Knowledge.
- •Get Detailed, Accurate, and Up-to-Date Answers.
- •Ask Questions Based on Your Personal Knowledge Base.
Supports context-based continuous search
Supports context-based continuous search and questioning
- •When you search for "what is hybrid ai search" for the first time
- •Continue to search for "advantages over google search" for the second time
- •MemFree will automatically rewrite the query to "what are the advantages of hybrid ai search over google search" based on the context.
Hybrid AI Search is Ready
Generate pages instantly using pre-designed, high-quality templates
- •Hybrid AI Search involves simultaneously performing searches on your personal knowledge base (such as bookmarks, notes, documents, etc.) and the Internet. This approach enables the retrieval of context that best matches your question, allowing a large AI model to provide the most accurate answer possible.
- •Personalization: The search results are tailored to your personal knowledge base, which is more relevant to you.
- •Efficiency: You can quickly find the information you need without having to search multiple sources.
- •Accuracy: The AI model can provide more accurate answers by considering both your personal knowledge base and the Internet.
- •Real-time Updates: Provides access to the latest information through Internet searches, ensuring timeliness and relevance
- •As shown in the figure, I first indexed the webpage https://www.memfree.me/blog/memfree-build-1-why
- •So when I ask Why Build MemFree, Hybrid AI Search will first consider the content of the memfree blog
- •then consider various content searched from the Internet.
AI Search based on Internet content
Today, MemFree supports AI Search and Ask based on Internet content.
Search the internet for the most relevant information by multi search engines
Search the internet for the most relevant images by multi search engines
Let LLM summarize the information and give you the best answer
Let LLM generate some related questions about the your question, You can keep asking questions until you get all the answers you like.