Elastic looks to speed up Jina Model Inference by 8x using NVIDIA RTX GPUs

Elastic is migrating its Jina AI models to NVIDIA RTX Pro 6000 Blackwell Server Edition GPUs to improve performance and reduce latency for agentic AI workloads. This transition has yielded significant results, including up to an 8x increase in embedding speed and 3.7x faster OCR conversion. Additionally, the new hardware ensures that small requests remain fast even when large, demanding jobs are being processed on shared GPU resources.