Nace launches Drex for Finance: 85.7% accuracy on 48 held-out finance benchmarks, +14.8 pts over GPT-6 Luna
testingcatalog · x · 2026-11-03
Nace AI announced Drex for Finance, a multimodal decision model post-trained for sub-second financial decisions like fraud checks, credit underwriting and compliance screening. It takes a document or transaction log plus a fixed answer set and outputs a probability score per option. Evaluated on 48 held-out finance benchmarks spanning 3,361 decisions, it scored 85.7% accuracy—14.8 points above GPT-6 Luna—and a 77.3 Finance Index vs 63.6 for Jev 1.14 and 65.8 for its own Drex-Max v1.1. The benchmark chart shows its biggest edge in document extraction (92.3 vs 48.8) and financial QA (96.7). Available on Nace Console.
More from Models
- GPT-6 edits Qwen 0.8B weights directly: 0.016% of parameters, 6x math gain, no training — teortaxesTex · 2026-11-03
- AI writes Breakout network weights in one reply, scoring 20× above PPO trained 500M steps — xennygrimmato_ · 2026-11-03
- DeepSeek's Underrated Role: Scaling and De-risking Ideas Academia Published Years Earlier — teortaxesTex · 2026-11-03
- Do faster quantized LLMs (Basalt, Infernix, Strata-nvfp4) trade away quality? — Kooky-Mode3047 · 2026-11-03
- OpenAI gives DevDay attendees 30 days of free Pro 100 tier — testingcatalog · 2026-11-03
- Peter Yang: Gemini beats ChatGPT at copy editing, but its harness holds it back — petergyang · 2026-11-03