Darrell Moore

I've tried the platform on the free tier, and I'm quite impressed with what it offers. I read a user saying it was a cheaper alternative to Manus. While I agree on the cost, the $20 monthly fee only gives you 5,000 points. I do complex searches for deep research, and each query uses about 500 to 750 points. Those 5,000 points won't last a month. But that seems to be the idea—just below the monthly price, they offer additional points you can buy in bundles. 5,000 extra credits cost $50, not another $20. Seeing the points per search was a bit disappointing. I understand that these searches and models are expensive to run and maintain, while still aiming to turn a profit. Google and OpenAI likely conduct their deep research queries at a major loss, which creates a false expectation of how many searches someone can do for $20 a month. So, while I’m not thrilled with the small number of points or how much each search consumes, I get why it’s set up this way. Hopefully, as models improve and prices drop—which has been happening—I can see the service either increasing monthly points or, more likely, reducing the points needed per deep search. I think the platform has potential, and many people will find real value in its analytical capabilities. It’s quite comprehensive. Kudos to the team, and I wish you great success. I also hope that economies of scale will allow you to offer more searches each month at the $20 rate without users needing to buy additional point packages. Bottom line: The searches are thorough, the analysis is accurate, and the reports are impressive, well-researched, and concise. They’re not bloated like Google and OpenAI, if that's what you prefer. PS. I considered docking one star for the price, but this is all new, and even the major players are figuring it all out. As far as the service is concerned, it's good, and if you can leverage this kind of research platform, you will get a lot of value from it.

I am updating this as of 2025/05/29. This Custom GPT now works for me. Perhaps it was an issue with my system at the time--I'm not sure. It 100% works and works well. I changed to five stars.

It's too bad that Pi.ai is no longer under active development, as the owning company is now focusing entirely on Enterprise models. :-( I enjoyed Pi, also. Its abandonment is part of what led me to find KIN. What were some of the recent developments you say that turned you away?

KIN AI COMPANION REVIEW (April 24, 2025 - v0.3.12) After extensive testing, here's my balanced take on this privacy-focused AI companion app. THE GOOD: Still completely free with strong privacy focus, mission-driven developers who actively seek feedback, helpful Discord community, great opportunity for early adopters to shape development, excellent local voice processing via Whisper protocol (or clodu-based Azure for older devices), constantly improving memory system that now references past conversations like ChatGPT, clean interface that doesn't try to do everything, regular updates with easy feature testing through settings. THE CHALLENGES: App stability still isn't production-ready (expect beta software quirks because it is), the in-app feedback system is clunky, though support via Discord works well, a major limitation is no web/desktop versions, and zero sync between platforms. Android and iOS devices act as separate instances with no shared data. BOTTOM LINE: Perfect for privacy-conscious users who don't mind beta testing and want to influence AI companion development. Skip if you need rock-solid stability or cross-platform sync. The principled team and active community make this worth watching, even if you must wait for more polish.

Update: As of May 29, 2025, it now works for me. (It no longer seems to work as of January 18, 2025.)