Paytm has quietly achieved a big milestone that needs attention. The company has become the first Indian firm to feature on the global capability leaderboard of the Spider 2.0 benchmark. It ranks in 2nd place worldwide with a score of 82.63.
It is important to note that Spider 2.0 is not a marketing-driven ranking. It is a serious benchmark presented at ICLR, one of the most respected AI research conferences globally. It tests how well an AI system can convert natural language questions into complex SQL queries across real, production-grade databases. This is exactly the kind of problem developers, analysts, and tech operations teams deal with every day.
To put it simply, this benchmark checks whether an AI can correctly answer business questions like “What changed last week?” or “Why did this metric drop?” using messy, real-world data. Many AI systems perform well in demos but fail in such environments. Ranking second globally here means Paytm’s system works where it actually matters.
Paytm achieved this using Prism, its in-house Text-to-SQL AI system. Prism does not rely on a single model, making decisions in isolation. Instead, it uses multiple AI agents that plan, validate, and refine queries together. This reduces errors and improves consistency, which is critical for financial and operational systems.
Most global AI leaderboards are dominated by US and Chinese companies or academic labs. Seeing an Indian consumer-tech company compete at this level shows how much the ecosystem has matured. It is also important to note that the company has developed this system internally.
For consumers, this kind of AI strength may not be visible on day one. But it directly impacts things like faster issue resolution, better fraud detection, cleaner insights, and more reliable digital services. When AI understands data better, products become smarter without becoming risky.
