Domain: https://ismy.blog Github:https://github.com/3289david/ismyblogYour blog. Nothing else.A minimal, self-hosted blogging platform. One blog per account, no post or visitor limits on the free plan. Pro is a single $25/year plan — AI writing tools, premium themes, custom CSS, advanced analytics, newsletter, redirects, webhooks, and API access.Stack Next.js 16 (App Router, Turbopack), React...
Domain: https://givemesometokens.dev Github: https://github.com/3289david/GiveMeSomeTokensGiveMeSomeTokens is a platform where supporters fund AI builders with Claude, GPT, Gemini, and OpenRouter and more API credits instead of dollars. Think BuyMeACoffee — but the currency is tokens, not coffee.Supporters connect their api key here and creators can use them with an open ai compatiable api,sdk...
Domain: https://hourly.krl.kr Github: https://github.com/3289david/hourlyFeatures: Time based web IDE Unlimited AI for that session A secure terminal with 0.5vcpu and 512MB ram Github integration Code EditorAccess your workspace on other devices too!hourly.krl.kr
ForgeLite is the most minimalized git. Repo:https://github.com/3289david/forgelite Features Repositories — public/private, fork, star, file browser, README rendering, commit history Issues — open, comment, close, Markdown body Pull Requests — real git diffs, code review comments, merge via git Git CLI — full HTTP smart protocol (git clone/push/pull works out of the box) Web Hosting — push to...
I made an unlimited free captcha service. I used altcha before, but it only focuses on Pow which was kind of hard to figure out bots. So,I made this service. It tests Pow+11 more stuff on your browser and updates every 10 minuites to stop new bots, is light, and easy to use! Maybe you should consider using this? There's a test for this: sh.krl.kr/test Hope this is great for you! Website: sh.krl.kr
It reminds me a bit of OpenAI Five — not just because it played a complex game, but because the real value wasn’t “AI plays Dota,” it was observing how coordination, strategy formation, and adaptation emerged under competitive pressure. A controlled RTS environment like this feels like a lightweight, reproducible version of that idea.
What I especially like here is that it lowers the barrier for experimentation. If researchers and hobbyists can plug different models into the same competitive sandbox, we might start seeing meaningful AI-vs-AI evaluations beyond static leaderboards. Competitive dynamics often expose weaknesses much faster than isolated benchmarks do.
Curious whether you’re planning to support self-play training loops or if the focus is primarily on inference-time agents?