Customer Reviews

Good morning again, sir. I’m not here to complain about absolutely anything—quite the opposite. I’m here to thank you. I had not found any other service operating under this kind of model that could support and save my projects the way yours has. I live in a country where it is extremely difficult to afford good AI services, mainly because the U.S. dollar is worth around five times the Brazilian real, and sometimes even six times as much. It is honestly incredible to me that the service you offer costs only $60. I just did the math, and based on my average usage of around 1.5 requests per second, I would never consume the full weekly allowance. I could probably even stay on the lower $33 plan. However, I intend to continue with the $60 plan, both to keep supporting your project and to give myself more room to experiment with sub-agents. Please keep me updated about your services and any future developments, because I genuinely intend to continue using them. And if you are able to maintain the $60 price for this amount of requests, I would be extremely happy. Thanks to this service, I will be able to complete the projects I am currently working on. Once again, I am extremely grateful. ----- I built this screen in my Agent Zero so I can see how much of the monthly quota I would have spent, if I were paying for a ChatGPT plan, to get the same response in terms of token usage — for the R$100, R$500, and R$1,000 plans. In less than a day of using it, it has already paid for itself several times over. I’m getting an AI that performs very well in coding and produces results comparable to ChatGPT 5.6. I’m happy for you to use these data for the company’s marketing or advertising if you’d like. Here in Brazil, however, the USD $200 plan is equivalent to about two-thirds of the average Brazilian worker’s base salary, while USD $60 is certainly much more affordable for a significantly better level of delivery. I’ll definitely recommend it to everyone close to me.

Hi my brother how are you doing today 😊 Kimi k3 is the best of all GLM is slow but good And I think my friend do big test yesterday and he have sheet for all modle

I’ve been using WE64 with Kimi K3 mainly for large Rust projects and agentic coding, and it’s been working really well for me. The 1M context and unlimited-token setup are especially useful for long coding sessions and large codebases. The API has also been reliable for me, including with very large requests, and the support has been helpful whenever I’ve had questions.