Nvidia and Reflection released open models companies can run on their own servers. Ramp and EY are already tuning Nvidia’s models for financial documents. Reflection launched with more than 40 partners, including AWS and Microsoft.
DeepSeek and MiniMax cut prices again. Both Chinese labs shipped updates that finish tasks faster for less. MiniMax claims a 70% cost reduction.
Inception and Perplexity bet on speed. Inception says Mercury 3 answers four to eight times faster than rivals. Perplexity’s new models make quick yes-or-no calls for routing and screening.
Chinese labs have led the open-weight artificial intelligence (AI) race for most of the year. This week, the U.S. stacked up its own entries, with Nvidia and Reflection releasing models built for enterprise work. DeepSeek and MiniMax responded with price cuts.
Complete the form to unlock this article and enjoy unlimited free access to all PYMNTS content — no additional logins required.
The open model market is no longer a Chinese monopoly. For banks and insurers that could not use Chinese models for security reasons, the choice of what to run in-house just got wider.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
Wyndham Deploys AI Agents to Boost Hotel Bookings and Cut Franchisee Costs
Source link







