European Financial Institutions Adopt NVIDIA Open Model Infrastructure
According to NVIDIA. NVIDIA announces new financial sector deployments leveraging accelerated computing and open-weight models.
- • Revolut is scaling its PRAGMA foundation model using NVIDIA Blackwell Ultra GPUs.
- • Stripe achieved 42x faster data pipelines and 92% cost savings in fraud detection testing.
- • European banks are utilizing open-weight models to maintain control over proprietary data.
Production-Scale AI Adoption
NVIDIA announced that European financial services organizations are building proprietary intelligence using open-weight models on accelerated infrastructure. Revolut is currently rebuilding its technology stack on the PRAGMA foundation model, utilizing NVIDIA cuDF and Blackwell Ultra GPUs to scale its production systems for millions of customers.
Operational Efficiency Gains
The publisher highlighted significant efficiency improvements across the sector. Neobank bunq reported that deploying on Nebius with NVIDIA accelerated computing matches the quality of leading closed models at 60% of the cost. Similarly, Stripe reported a 42x speed increase in data pipelines for its fraud-detection platform during offline testing.
Advanced Financial Research
NVIDIA detailed how firms like LSEG and RavenPack are utilizing its hardware to accelerate research. LSEG is working to analyze historical tick data up to 10x faster using the Polars GPU engine, while RavenPack is using fine-tuned Nemotron models to automate the generation of backtest-ready datasets for market trend analysis.
- 1. blogs.nvidia.comOpen source