This AI Does Not Exist
Hey HN! Author of the site here. I tried a few tricks to keep the text-generation part of the site up, but even leaning hard on Huggingface's API and bumping time-outs up, it looks like the site is struggling a bit. I'm going to see if there's anything I can do to keep the text-generation part available, but in the meantime, the pre-generated set should stay pretty stable. Not sure if there's much else I can do without burning a hole in my cloud bills — sorry for the troubles! I've put up a more detailed description of how this works on the GitHub -…
In plain words
This AI Does Not Exist generates realistic text descriptions and images of fictional people, places, and things that do not actually exist. It uses machine learning models to create novel content on demand, with a library of pre-generated examples available when live generation is unavailable. The site is designed for users curious about AI-generated content and how generative models work, offering both real-time generation through Huggingface's API and a stable collection of previously created samples.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN! Author of the site here. I tried a few tricks to keep the text-generation part of the site up, but even leaning hard on Huggingface's API and bumping time-outs up, it looks like the site is struggling a bit. I'm going to see if there's anything I can do to keep the text-generation part available, but in the meantime, the pre-generated set should stay pretty stable. Not sure if there's much else I can do without burning a hole in my cloud bills — sorry for the troubles! I've put up a more detailed description of how this works on the GitHub - https://github.com/thesephist/modelexicon PS - if anyone at Huggingface is reading this and wants to help out with keeping the API up, that would be super :)
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, April 2022
the whole month →- WA
Hi HN community, I’m Zach, founder and CEO of Warp, and am excited to show you Warp, a fast Rust-based terminal that’s easy to use and built for teams. As of today, Warp is in public beta and any Mac user can download it. It works with bash, zsh, and fish. The terminal’s teletype-like interface has made it hard for the CLI to thrive. After 20 years of programming, I still find it hard to copy a command’s output; I always forget how to use `tar`; and I always have to relearn how to move my cursor. To fix fundamental accessibility issues, I believe we need to start innovating on the terminal,…
Dev tools · 2022 · warp.dev
- SS
Life & fun · 2022 · slowsocial.us
- MY
Implement scanyourpdf.com in JavaScript. No backend servers needed.
Life & fun · 2022 · lookscanned.io
