Chessdream – Generate realistic chess positions with AI
For no specific reason, I trained a neural network to generate random chess positions that look similar to positions from actual games (from lichess db). I also made it so you can condition it on some fixed pieces, and adjust the number of pieces. It turned out to be quite effective and I find it surprisingly fun and instructive to generate e.g. endgame positions with a certain pawn structure (set low temperature, place some pawns and position the kings, adjust number of pieces to get an endgame), and then figure out how to win vs. the computer in those positions. I hooked it up so that you…
In plain words
Chessdream uses a neural network trained on real chess games to generate realistic positions, allowing users to set specific pieces, adjust piece counts, and control randomness. Designed for chess players seeking practice, it lets users generate custom endgames, openings, or themed positions, then play against computer opponents. The tool combines position generation with built-in analysis through lichess integration and direct play via the Noctie engine, making it useful for studying specific pawn structures and tactical scenarios.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
For no specific reason, I trained a neural network to generate random chess positions that look similar to positions from actual games (from lichess db). I also made it so you can condition it on some fixed pieces, and adjust the number of pieces. It turned out to be quite effective and I find it surprisingly fun and instructive to generate e.g. endgame positions with a certain pawn structure (set low temperature, place some pawns and position the kings, adjust number of pieces to get an endgame), and then figure out how to win vs. the computer in those positions. I hooked it up so that you can easily play vs. a computer, either via lichess analysis → continue from here, or via my own project Noctie. I’m thinking about whether I could develop it further to create position variants on a certain theme that the AI thinks I need to practice, or maybe make a PvP feature where you play a random position vs. a human. Ideas or feedback?
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 · 27d ago · cactuscompute.com
