Nace AI has published a 51-page report on how it trained Drex, its family of open-weight…
This is a dev post classified by AGK Intelligence as AI dev tools (a launch), kept by the Dev Radar because it carries real work, not commentary.
Nace AI has published a 51-page report on how it trained Drex, its family of open-weight AI models that pick the most likely answer from a fixed list of options. Drex v1.5 was trained on 1.15 million example decisions with verified answers, covering tasks from reading contracts to spotting phishing emails. In a second stage, a larger AI model taught Drex how confident to be in each answer, so its percentages match how often it is actually right. Drex can be tested in Nace Console, where users upload documents with questions and receive a probability score for each answer.
Posted by 🚨 AI News | TestingCatalog (83.4k followers) 1 h ago · 67 likes · 5.7k views · view the original post on X. Kept by the Dev Radar as AI dev tools. Tools mentioned: Console by Nace.AI.
More dev work like this
- The world of crypto is a bit shaken right now but @reppo, we haven’t stopped shipping 🚢 — @reppo
- 🔌 If you're using Devin on an 𝝠 Expo app, there's an official plugin for it now. — @expo
- Picking an AI agent stack shouldn’t start with 30 random tabs. — @DanKornas
- Meet Qwen3.5-4B-3D-Coder: a 4B model fine-tuned to turn text into Three.js 3D scenes.… — @HuggingModels
- TRENDING 📈 @elliotarledge reverse-engineered Wispr Flow and used DeepSeek to fine-tune… — @thestartpod
- EVAL (0x10f52295e129817e8e4b28da0dab067321ec2222) staking is now live on… — @reppo
- I used Composio Instant with Claude Code to check how my own product, Krevox, shows up… — @neerajjj6785
- Zyphra is building something interesting called MAIA. — @TeksEdge
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.8k posts from 5k X accounts over the last 21 days, 9.5k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 20:12 UTC. Full method.