@spec_tacul_ar · X

Specifications for humans and AI.

@ErnestPresident · X

I know how to code Sun Yi so I vibe coded a multi layers math pipeline + multi layer verifier pipeline that do advanced calculus, I must do something wrong then.

cdnsteve · HN

Tokenstead, find AI models for your hardware

Humanizadordeia.app — 将AI文本转化为自然人类写作。1）人性化处理，让 AI 文本更自然、更贴近人工写作与品牌口吻；2）检测服务，帮助识别潜在 AI 生成内容，提升合规与质量。支持多语种、极速处理、保留语义与风格。免注册体验 300 字，适合博主、营销团队、跨境站点、教育与企业内容运营。以先进模型（GPT、Claude、Gemini 等）驱动。

@EluraCode · X

Check out what I just built with Lovable!

ai world generator — 用文本和照片生成自然风光视频的 AI 工具网站

@salmanzafar949 · X

@mageofweb3 · X

@victorvravatar · X

white_tiger · HN

Physicsbase.ai Validated Physics for AI

@fedeponte1 · X

my first saas. i think i built something people actually want and completely failed at the business side. would genuinely love help. openpaper ( writes full, cited research papers from a prompt. research, outline, writing, real citations, export. the honest numbers: - ~2,300 signups - ~1,500 generated a full paper. people actually use it. - ~5 have ever paid. ~$45/mo. activation is great, revenue is basically zero. i've been staring at this for weeks. the paywall is on export (pay ~$12 to download the pdf). but the full cited paper is right there on screen, free, you can read it and copy-paste it. so i'm charging for the gift wrap after handing over the gift. almost nobody even reaches checkout. 76% generate exactly one paper and leave. one-and-done, so subscriptions make no sense. the audience is mostly students and researchers on personal gmail, all over the world, almost none on university or company emails. traffic is x and github. so i genuinely can't tell

fran-mora · HN

I gave 5 AI coding agents one prompt: grow a banana plant through its whole life in three.js: sprout, leaves, flower, fruit, rot, then pups that restart the loop. It's deceptively simple and yet very hard to get right from procedural code: you have to write working three.js and understand how the plant is actually built; how it hangs, ages and decays. Get the biology wrong and the code renders something weird. These are agents, not bare models (Claude Code and Codex for now). They can use tools, including playwright to check their work and improve it.