Case Study: Tarteel AI achieves 22% lower latency and 56% lower costs with CoreWeave

Tarteel Migrates Cloud Infrastructure to CoreWeave with Help from Zeet

Tarteel AI, the AI-powered Quran study app, needed a better cloud setup for its growing real-time voice and speech workloads. After finding AWS limited in GPU options, observability, and production inference optimization, Tarteel turned to CoreWeave, with support from Zeet, to support its AI deployment needs.

CoreWeave provided Tarteel with access to NVIDIA Riva and NeMo, plus a broader range of NVIDIA GPUs, scalable infrastructure, and better observability. Using CoreWeave and Zeet, Tarteel completed its migration in just 1–2 days and achieved a 22% improvement in latency, median request latency of 423.5 ms, and about a 56% reduction in costs.

Tarteel AI

Anas Abou Allaban

Co-Founder and CEO