Upend: The New AI Search Engine Supporting 100+ LLMs

Updated: May 08 2024 00:22

A new player has emerged from stealth mode to challenge the likes of Google and Microsoft in the realm of AI-powered search. Upend, a Canadian startup founded by Jeevan Arora, a student from the University of Toronto School, aims to democratize access to advanced generative AI models by offering an affordable and comprehensive AI search platform.



What started as a summer project by Arora quickly gained traction, prompting him to develop Upend into a full-fledged platform for enterprise teams. With a mission to make advanced generative AI models more accessible and affordable, Upend has the potential to disrupt the AI search market, currently dominated by players like Perplexity.

• Extensive Model Library: Upend boasts an impressive collection of 100 large language models (LLMs), including both general-purpose models from OpenAI, Claude, and Mistral, as well as task-specific models like Meta's Code Llama and Deepseek Coder. This extensive library sets Upend apart from competitors like Perplexity, which offers only a handful of mainstream LLMs.


• Grounded Answers with Citations: Users can select any LLM from the available options and ask questions related to work or everyday life. The chosen model combines the query with web search or selected sources to provide grounded answers, complete with citations for easy reference.


• Wikipedia-based Answers: Upend offers the unique option to base answers on Wikipedia, a feature not currently available on Perplexity.


• Affordable Pricing: Upend's team plan is priced at $20/month, providing access to all models in its library up to a certain token threshold. The company claims to offer up to 90% cost reduction compared to competitors like Perplexity and ChatGPT.


• Student-friendly Pricing: Recognizing the importance of making AI accessible to students, Upend offers a discounted student plan at $5/month.

While Upend is still in its early stages and has room for improvement, particularly in terms of user experience, Arora has ambitious plans for the platform. He envisions Upend evolving into a full-fledged task engine that not only provides answers but also executes tasks for users. For example, if a user searches for "How to convert my PNG image to JPG," Upend would provide an answer and offer to complete the task on their behalf.

In the near future, Upend plans to expand its model library, introduce single-sign-on (SSO) for easier team onboarding, and explore the development of a mobile app and API.


The rise of AI-powered search engines like Upend and Perplexity is expected to have a significant impact on the traditional search market. Gartner predicts that traditional search engine volume will drop by 25% by 2026, with search marketing losing market share to AI chatbots and virtual agents. As generative AI solutions become more embedded across all aspects of the enterprise, companies will need to rethink their marketing channels strategy.

Upend's emergence from stealth mode marks an exciting development in the world of AI-powered search. With its extensive model library, affordable pricing, and ambitious roadmap, Upend has the potential to democratize access to advanced generative AI models and challenge established players in the market. As the AI landscape continues to evolve, it will be fascinating to watch how Upend and other innovative startups shape the future of search and task automation.

Below is the list of LLMs models that Upend currently support: