The increasing adoption of generative AI in enterprises has led to a growing demand for trusted, performant, and cost-effective solutions. IBM’s Granite code models, developed by IBM Research, are now available on the NVIDIA API catalog as NVIDIA-hosted NIM inference microservices. This collaboration between IBM and NVIDIA aims to drive enterprise gen AI adoption by pairing NVIDIA AI Enterprise software and accelerated computing with industry solutions from IBM Consulting. The Granite code models, optimized for higher throughput and performance, are designed to simplify and accelerate the deployment of AI models across GPU-accelerated workstations, data center, and cloud platforms. With their availability on the NVIDIA API catalog, enterprises can easily use industry-leading models for trusted code generation and translation, GPU infrastructure, and inference management software capabilities for price-performance optimization.

IBM Granite Code Models Now Available on NVIDIA API Catalog
IBM Granite code models can outperform some models that are even twice their size, according to HumanEvalPack evaluation.










