Orpiva

ORPIVA is a London-based AI startup helping brands to deliver marketing campaigns like no other. Its platform creates high-performing marketing content at scale and low cost, empowered by trend-driven influencers and Generative AI. In addition to creating immersive influencer content, its advanced Generative AI seamlessly integrates product placements into influencer trend videos, streamlining the video licensing process for its clients at scale with a single click. ORPIVA’s paid and organic content consistently outperforms average social media content by 500%. It has garnered over 100 million views across various social media platforms for renowned brands like Samsung, Vodafone, Boohoo, and Nike.

To develop its AI-powered applications, ORPIVA was using a range of hosted GPU solutions from multiple cloud providers, alongside its own on-premises servers. However, managing and maintaining all these solutions was becoming costly and burdensome, leading to inefficiencies. As a fast-growing startup, ORPIVA needed to simplify its cloud estate and find a partner able to deliver high-performance GPUs on a reliable, secure and affordable cloud. This would enable the company to scale its AI platform to meet customer demand and avoid wasted expenditure.

ORPIVA found OVHcloud to be the logical choice due to its simple, powerful, and cost-effective cloud solutions. The two companies partnered and ORPIVA adopted NVIDIA Tesla V100 GPU instances hosted on OVHcloud’s Public Cloud platform.

NVIDIA Tesla V100 GPU instances deliver unparalleled performance for businesses running AI and Machine Learning processes. Boasting the power of 100 Central Processing Units (CPU) within a single GPU, these GPUs mark the first to surpass the 100 TeraFLOPS (TFLOPS) barrier in AI performance. This supports rapid training of AI models, enabling businesses to go-to-market significantly faster.

As these GPUs were hosted on the Public Cloud, ORPIVA benefitted from a reliable cloud infrastructure at a 99.99% Service Level Agreement (SLA). Being spread across multiple global data centres, the Public Cloud ensures high-availability and redundancy, and offers several performance-enhancing features, such as Load Balancers. ORPIVA leveraged this to distribute traffic and handle peak usage more easily, thus reducing downtime and ensuring a consistent service for its customers. The Public Cloud also offers multiple security features, such as Anti-DDoS Protection, firewalls, and encryption. These are crucial for AI companies like ORPIVA, which must ensure compliance with data protection laws.

Finally, OVHcloud’s transparent pay-as-you-go billing gave ORPIVA greater control over its cloud expenditure, enabling the company to avoid the additional hidden costs and vendor lock-in often associated with other cloud providers.

Switching to OVHcloud’s comprehensive Public Cloud solution enabled ORPIVA to streamline its processes, cut costs, and increase efficiency. Being based on a resilient and secure infrastructure, the Public Cloud reduced the risk of downtime, providing peace of mind for ORPIVA as a growing startup with an enterprise customer-base. Leveraging the capabilities of NVIDIA GPUs on the Public Cloud enabled ORPIVA to train its AI solutions much faster, significantly reducing the training time of its language and video synthesis models by 15.2% and 35.7% respectively.

Thanks to OVHcloud’s unambiguous pay-as-you-go model, ORPIVA was also able to reduce its monthly cloud costs. This combination of performance, simplicity and transparent billing enabled ORPIVA to scale much faster and more cost-effectively compared with its previous cloud providers, giving the company a significant competitive advantage in a dynamic market.

“We have been using OVHcloud solutions to develop and deploy our AI-powered applications, including LLM training and automatic AI video generation. We are very satisfied with their performance, reliability, and support.”
Salman Valibeik, CEO and Co-Founder at ORPIVA