... released 3 models. Atome LM : Atome LM, an AI that runs on A $5 ESP32. No internet, no os, no data center. Tilelli LLM : A language model that say's "i don't know" instead of bluffing. Yaz KB Model : Unlike traditional AI, you can directly add, read, update, or delete facts in its memory like a simple database without needing to retrain the model. What do you think ...

... noticed something strange when working with AI like ChatGPT and Gemini. You can craft the most elegant prompt in the world, but once the conversation runs long, the model quietly forgets what was said earlier. It starts bluffing, filling gaps with confidence, like someone trying to recall a story they only half remember. That made me rethink what prompt engineering even is. Maybe it s not just about how you start a conversation, but how you keep it coherent once ...

... versioned memory + rollback so agents actually stay grounded. check it here- github.com/InfinitiBit/graphbit Comment from Ghost Kitty(@ghost-kitty-7815048): Comment Deleted Comment from Sanskar Yadav(@sanskarix): If models tracked their own uncertainty and flagged it (instead of just bluffing), trust would go way up. Curious to see how versioned memory plays out in practice ...

... build your own UI (Or ask us we can help you build custom UI inside your stack). Let me know if you want to get a demo. Comment from Rohid(@rohid): @bdzhel  Great, now we are competitors! Comment from Rohid(@rohid): @bdzhel  I looked at CoSuport AI. It's cool. We are also focusing on logic, better AI response. Also our main goal is to make AI accessible from anywhere in your website. So we focus both ... ... interaction. Comment from Edward Look(@edlook): I think the trust has to be built over time. Maybe a "human approval" step for the AI-drafted responses could be the bridge. Comment from Marcus Freeland(@marcusfreeland): @bdzhel   Bluffing