26:34
I Ranked Every AI Tool (May 2026)
AI Edge
Overview
This video provides a ranked list of various AI tools the presenter has personally used, categorized into tiers from S (can't live without) to F (complete skip). The presenter explains their reasoning for each ranking, highlighting strengths, weaknesses, and use cases, with a focus on practical application for users. The list includes LLMs, coding assistants, design tools, voice generators, video creation platforms, and memory systems, emphasizing how the rapidly evolving AI landscape impacts the utility and value of these tools.
How was this?
Save this permanently with flashcards, quizzes, and AI chat
Chapters
- AI tools are ranked into tiers: S (essential daily), A (daily driver), B (power tool for specific tasks), C (situational), D (overhyped/not worth it), and F (avoid).
- The ranking is based on personal experience and practical daily workflow integration.
- The video aims to identify the best AI tools worth the money and highlight underrated 'sleeper picks'.
- The AI landscape is rapidly changing, necessitating frequent updates to such rankings.
Understanding this ranking system helps viewers interpret the presenter's evaluations and apply them to their own needs and priorities when choosing AI tools.
The S tier is defined as tools the presenter 'can't live without' and uses daily, while F tier tools are recommended to 'just avoid'.
- Cursor, a coding interface with AI, is ranked B (power tool) as Claude Code has improved significantly, reducing its unique advantage.
- Manis is a 'sleeper pick' ranked A (daily driver) for its effectiveness in agentic workflows, allowing automated tasks and efficient token usage.
- Notion AI is ranked C (situational) because it relies on other LLMs (GPT, Opus, Gemini) and is only truly useful for existing Notion users.
- Zapier and N8N, once essential for automation, are now ranked D (overhyped) as LLMs like Claude can handle many of their previous functions directly.
This section highlights how core functionalities like coding and automation are being integrated into or surpassed by more versatile LLMs, changing the necessity of specialized tools.
Manis can schedule tasks to run automatically at a specific time, like 4 a.m. every night, demonstrating its agentic capabilities.
- Llama (Meta's open-weight model) is ranked D (overhyped) due to better open-source alternatives and high hardware requirements for local inference.
- Granola is ranked B (power tool) for meeting notes due to its local privacy and integrated summarization, preferred over the more known Fireflies (C tier).
- NotebookLM is ranked A (daily driver for students, power tool for others) for its utility in transcribing, summarizing, and learning about topics, especially within the Google ecosystem.
- Replet, a browser-based IDE with AI capabilities, is ranked C (situational) as Claude Code has diminished its unique advantage for building AI apps.
- Versel is ranked A (daily driver) for deploying AI applications and websites, offering significant customizability and ease of use, especially with its agent.
This chapter explores tools that enhance productivity and development workflows, showing how integrated AI features and platform ecosystems influence their ranking.
NotebookLM is particularly useful for students needing to transcribe YouTube videos or summarize complex topics for study.
- Lovable, a tool for building AI websites/interfaces, is ranked C (situational) as Claude Design has become a strong alternative, though Lovable offers a cleaner UI for specific tasks.
- Midjourney is ranked F (complete skip) due to better, more user-friendly, and accurate image generation models now available.
- Canva's AI features are ranked C (situational) for designers, especially its Magic Studio, though it relies on other LLMs for some functions.
- 11 Labs (voice cloning/generation) and HeyGen (AI avatar/video generation) are both ranked B (power tool) for their specialized, high-quality output in media creation.
- Sunno, an AI music generator, is ranked C (situational) despite its impressive output and potential impact on music production, due to its niche application.
- Runway, an AI video editing studio, is ranked C (situational) as Higsfield offers a more comprehensive video solution.
- Cans (video model) is ranked C (situational) but noted as the go-to for pure video generation if not using an all-in-one platform like Higsfield.
This section covers tools for creative professionals, demonstrating how specialized AI in areas like voice, video, and music is evolving and being consolidated into broader platforms.
HeyGen is highlighted as the best tool for creating AI avatars and translating videos.
- Higsfield is ranked A (daily driver) as an all-in-one video AI platform, integrating multiple top video models (e.g., Cans, Google VO) and offering agentic features for content creation.
- Perplexity is ranked A (daily driver) for its ability to orchestrate various LLMs, provide automated daily market updates, and offer a powerful personal computer for advanced use.
- Superbase is ranked A (top AI tool not being used) as a portable, cloud-based memory system that syncs user data across LLMs, ensuring data sovereignty.
- Firebase is ranked C (worse than Superbase) as Google's backend service, lacking the data ownership and AI integration benefits of Superbase.
- Super Memory is ranked as a potentially easier alternative to Superbase for data enrichment and auto-classification but lacks full data ownership.
These tools represent the trend towards integrated platforms and robust memory solutions, which are crucial for managing and leveraging AI effectively across different applications.
Superbase allows users to maintain a consistent memory across different LLMs, preventing data loss when switching tools.
- The Chinese open-source models Deepseek (A tier) and Quen (B tier) are highlighted for their strong reasoning and cost-efficiency compared to US models.
- Grok is ranked C (situational) for its strength in real-time news querying but lacks broader capabilities compared to top LLMs.
- Gemini is ranked A (daily driver) due to its deep integration into the Google ecosystem and constant updates, despite not being the absolute leader in every category.
- GPT (specifically GPT-4/5) is considered the smartest and most trustworthy model for strategic tasks and complex coding, but lacks feature richness compared to Claude.
- Claude is ranked slightly higher than GPT (both could be S tier) due to its superior creative capabilities, design integration, and overall usage frequency by the presenter.
- The presenter acknowledges that the 'best' tool is subjective and depends heavily on individual user needs and workflows.
This chapter focuses on the leading Large Language Models and emerging open-source contenders, illustrating the competitive landscape and the trade-offs between raw intelligence, features, and cost.
The presenter uses GPT for strategic planning and complex coding tasks where absolute accuracy and trustworthiness are paramount.
Key takeaways
- The AI tool landscape is highly dynamic, requiring continuous evaluation and adaptation of tool choices.
- Specialized AI tools are increasingly being integrated into or surpassed by more versatile, all-in-one platforms and advanced LLMs.
- Data sovereignty and portable memory systems (like Superbase) are becoming critical for users who switch between different AI tools.
- Cost-efficiency is a significant factor, with open-source models from China offering competitive performance at a lower price point.
- While top LLMs like GPT and Claude excel in different areas, the choice between them depends on whether creativity/integration or raw strategic power/trust is prioritized.
- AI tools that enhance productivity, automate workflows, and facilitate creative output are valuable, but their ranking is highly dependent on individual user needs and existing ecosystems (e.g., Google, Notion).
- Understanding the specific strengths and weaknesses of each AI tool is crucial for maximizing its utility within your personal or professional workflow.
Key terms
AI Tiers (S, A, B, C, D, F)Agentic WorkflowsLLM (Large Language Model)Open-Weight ModelLocal InferenceData SovereigntyMemory SystemIDE (Integrated Development Environment)Multimodal AIUGC (User-Generated Content)
Test your understanding
- How does the presenter's tier system (S-F) help in understanding the practical value of different AI tools?
- Why are tools like Zapier and N8N being de-emphasized in favor of more integrated LLM capabilities?
- What makes Superbase a critical tool for users who frequently switch between different AI models, and why is data sovereignty important?
- Compare and contrast the strengths of Claude and GPT according to the presenter, and explain why one is ranked slightly higher for their personal use.
- How do factors like cost-efficiency and integration into existing ecosystems (e.g., Google) influence the ranking of AI tools like Deepseek and Gemini?