Understanding Liberty Modes: Search, Visual, Reasoning & More
Learn what each Liberty mode does and when to use them. From web search to code analysis, master all available modes for optimal results.
Tanvir A
Co-Founder & Senior AI Engineer
8 min read
Liberty provides multiple modes, each optimized for different types of tasks. Think of modes as specialized tools - you wouldn't use a hammer for a screw. Similarly, selecting the right mode ensures you get better results with lower latency and cost.
Available Modes at a Glance
Liberty supports the following modes. Each has distinct capabilities and performance characteristics:
Chat: Standard conversational AI, general-purpose tasks
Search: Real-time web search, current events, up-to-date information
Visual Studio: Image generation, image edit, video generation
Study & Learn: Deep thinking, complex problems, step by step explanations
Extreme: Extreme research , find information by searching more than 200 sources
Mode 1: Chat Mode
Chat Mode is standard conversational AI for general-purpose tasks and everyday questions.
Best For: General questions, conversation, brainstorming, everyday assistance
Speed: 2-5 seconds
Example Query: What's a good recipe for pasta?
Mode 2: Search Mode
Use Search Mode when you need current information, web results, or real-time data.
Search Mode Workflow
You Input Query
Liberty Searches Web
Aggregates Results
You Input Query
Liberty Searches Web
Aggregates Results
Analyzes 20+ Sources
Synthesizes Answer
Cites Sources
Analyzes 20+ Sources
Synthesizes Answer
Cites Sources
Best For: Any information on the web, documents, stock prices, news, weather, product reviews, event schedules
NOT For: Image generation, casual chat
Speed: 3-8 seconds (includes web crawling) when TEE is disabled; 10-60 seconds when TEE is enabled
Example Query: What are today's Bitcoin price movements?
Mode 3: Visual Studio Mode
Visual Studio Mode enables image generation, image editing, and video generation.
Best For: Creating images, editing visual content, generating videos
Capabilities: Image generation, image edit, video generation
Speed: 5-15 seconds depending on complexity
Example: Generate a futuristic city landscape or edit an existing image
Mode 4: Study & Learn Mode
Study & Learn Mode activates deep thinking - the LLM spends more time on complex problems, reasoning step-by-step through their logic for comprehensive explanations.
Best For: Math proofs, logic puzzles, strategic planning, complex analysis, step-by-step explanations
Speed: 2-60 seconds
Example: Complex physics problem → Model shows step-by-step derivation and explanation
Mode 5: X/Twitter Mode
X/Twitter Mode is optimized for searching across the X/Twitter platform and summarizing information.
Best For: Searching and summarizing information from X/Twitter platform
Speed: 5-20 seconds depending on query complexity
Example: What are the latest trends in AI on X/Twitter? → Gets relevant tweets and summarizes key points
Mode 6: Research Mode
Research Mode queries academic databases and peer-reviewed sources for citations and scholarly information.
Best For: Literature reviews, academic writing, scientific fact-checking
Speed: 5-10 seconds (searches academic databases)
Example: Summarize recent advances in transformer architectures → Gets recent papers with citations
Mode 7: Reddit Mode
Reddit Mode enables real-time Reddit search and summarization for trending discussions and community insights.
Best For: Finding community perspectives, trending discussions, Q&A from Reddit communities
Speed: 5-20 seconds (searches Reddit communities)
Example: What do developers think about Rust? → Gets relevant Reddit discussions
Mode 8: YouTube Mode
YouTube Mode analyzes video content for summarization and information extraction from YouTube videos.
Best For: Video content analysis, summarizing long videos, extracting key information
Capabilities: Transcript analysis, key points extraction, video summary generation
Speed: 5-60 seconds depending on video length
Example: Summarize a 1-hour tutorial → Extracts key concepts and timestamps
Mode 9: Extreme Mode
Extreme Mode enables extreme research - searching more than 200 sources for comprehensive, in-depth information discovery.
Best For: Deep research, finding obscure information, comprehensive analysis across many sources
How It Works: Searches 200+ sources (web, academic, news, forums) and synthesizes findings
Speed: 30-300 seconds (searches many sources)
Example: Comprehensive history of a technology → Finds information across Wikipedia, academic papers, forums, and blogs
How to Switch Modes
Switching modes in Liberty is simple:
Locate Mode Selector: Chat input shows a mode selector with dropdown
Click Dropdown: See all available modes with descriptions
Select New Mode: Click your desired mode - UI updates immediately
Continue Conversation: Previous context is preserved when switching modes
Pro Tip
You can switch modes mid-conversation. Starting with Search Mode for research, then switching to Study & Learn Mode to analyze findings is a powerful workflow.
Mode Selection Decision Tree
Use this flowchart to quickly decide which mode to use:
Which Mode Should I Use?
General conversation?
YES
Chat Mode
General conversation?
YES
Chat Mode
Need current info?
YES
Search Mode
Need current info?
YES
Search Mode
Generate/edit images?
YES
Visual Mode
Generate/edit images?
YES
Visual Mode
Need deep thinking?
YES
Study & Learn Mode
Need deep thinking?
YES
Study & Learn Mode
X/Twitter Trends?
YES
X/Twitter Mode
X/Twitter Trends?
YES
X/Twitter Mode
Academic writing?
YES
Research Mode
Academic writing?
YES
Research Mode
Want Reddit insights?
YES
Reddit Mode
Want Reddit insights?
YES
Reddit Mode
Analyze videos?
YES
YouTube Mode
Analyze videos?
YES
YouTube Mode
Extreme research?
YES
Extreme Mode
Extreme research?
YES
Extreme Mode
Did you find this helpful?
Share with others exploring privacy and decentralized AI.
