Nace AI

How are Decision Models Trained?

Today, we are publishing the Drex Technical Report for the developer community! 🚀

Discover details on how the State-of-the-art Open-Source decision models are trained with RLAF.

Build more at: console.nace.ai (try for free)

P.S. Drex Max 1.1 and Drex for Finance are closed sourced and designed for enterprise use-cases

Drex Max v1.1 is #1 on Decision Index @ArtificialAnlys and going open-weight soon!

Both achieving SOTA and Pareto frontier, it’s one of the fastest & most capable Decision models today

Drex API: nace.ai/drex (~1M free requests)
Paper: arxiv.org/pdf/2611.07854

Make better decisions.

Qwen 4 today

Drex is officially open source. 🧬

Now come evolve something with it!

Drexathon: Nov 14, San Francisco. 5 hours to build agents that don't just chat, they decide. 1st place takes a Mac Mini!! 🏆

Don't go extinct 🦕
luma.com/nace-drexathon

Announcing Drex for Finance - the multimodal decision model, post-trained for sub-second financial decisions.

Financial operations require precise, regulatory-grade decisions with zero hallucinations across filings, support, credit risk, live fraud prevention, quant signals, and compliance.

Evaluated on 48 held-out finance benchmarks:
- 85.7% accuracy across 3,361 decisions (+14.8% vs. GPT-6 Luna)
- 77.3 Finance Index (+13.7 vs. Jev 1.14, +11.5 vs. Drex-Max v1.1)

Trained for sub-second financial intelligence across the entire pipeline:
- Financial Documents & Filings: Extraction decisions, classification, and clustering
- Credit & Lending: Automated underwriting, PD modeling, and EWS
- Fraud & AML: Sub-second transaction scoring, ATO defense, and SAR clustering
- Markets & Trading: Regime detection, meta-labeling, and LOB execution signals
- Insurtech & RegTech: Claims fraud detection, sanctions screening, and audit reconciliation

Available exclusively on Nace Console: console.nace.ai