newest first
all setups
Every setup on this site, with the numbers. Most of them are the write up behind a reel.
- A freelance proposal that opens with their problem, in one paste (the PITCH setup)
- The photos you deleted are still there (the PHOTOS steps)
- The scam call that sounds like your bank, and the two rules that beat it (the SCAM page)
- Three money jobs you keep putting off, done in one paste each (the MONEY setups)
- That three minute voice note, read for you (the VOICENOTE setup)
- Split any trip, rent or dinner bill fairly, in one prompt (the SPLIT setup)
- Your weekly AI research, done while your laptop sleeps (the RESEARCH setup)
- Serve a 1.02T open model: the MiMo-V2.6-Pro setup
- Use MiMo-V2.6-Pro from OpenRouter or Command Code, no GPUs
- This week's dinners and the shopping list, from one fridge photo (the MEALS setup)
- ChatGPT dots: who gets one, how to set it up, and the settings to check first
- 209 comments sorted in 21 seconds: the exact Jev setup
- Try my AI trading bot free, on fake money (the BOT setup)
- Your bank statement, sorted by category in one paste (the SORTED setup)
- Pick the right GGUF for your RAM (the quant picker)
- Nine agent skills trending this week, plus a SKILL.md starter
- Run Microsoft BitNet on your CPU
- Keep your AI assistant running when your laptop is closed
- ElevenLabs $100, your GPU 6 cents? The exact VoiceStudio setup
- Nine harnesses, one model, and a 17x spread in the bill
- Run a 125B model on a 48GB laptop (slotstream setup)
- Fine tune an 8B model on a 4GB laptop GPU (Soup setup)
- Train your own LLM for 43 cents and 2 hours (minimind setup)
- Send your AI pictures of your code instead of the code (pxpipe setup)
- Give your coding agent a throwaway Linux VM (clawk setup)
- Rebuild Redis from an empty file (Ship That Code setup)
- 7.4 billion free AI tokens a month
- Run a 20B AI model in 7GB RAM
- Run a 35B model on a 300 euro computer (CarWatch setup)
- Run a 27B model in 6.2GB, no GPU (Qwen3.8 setup)
- Run a 744B model in 9.9GB of RAM (Colibri setup)
- Run a 30B model on a 2016 desktop (quantprobe setup)
- Run a 2.78 trillion parameter model in 8GB (Kimi K3 setup)
- Run an 80B model in 4.3GB of RAM (Swiftlet setup)
- Run AI on a laptop with no graphics card
- Run a 14MB AI on a Raspberry Pi (Needle2 setup)
- Run a 70B model on a 4GB card (AirLLM setup)
- The weird web list
- How I Built My JARVIS