Global artificial intelligence (AI) search startup "Liner" announced on the 25th that it achieved a score of 93.7 points on the SimpleQA Benchmark, proving its top-tier performance.
This is the highest score among AI search engines and large language models (LLM) in the same category, demonstrating that Liner's AI search technology has gained competitiveness in the global market, according to the company.
SimpleQA is a benchmark that evaluates how accurately AI can find facts, measuring the response accuracy of AI based on 4,326 questions across various fields, including science, politics, popular culture, and art. In this test, Liner achieved the highest score, surpassing Perplexity Pro (90.6 points), Gemini 2.0 Flash with grounding (84 points), and GPT-4o (38.4 points).
Liner's recent achievement is based on advanced query generation technology and a sophisticated search process design. It maximized the reliability of AI search by generating optimal queries and prioritizing the retrieval of documents containing accurate information. In particular, it showed strengths in providing users with accurate information by minimizing hallucination.
Kim Jin-woo, CEO of Liner, said, "Liner already possesses AI search technology that surpasses global big tech corporations, and this achievement once again proves our technological edge," adding, "We will evolve into an AI search service that provides accurate information to users based on trustworthy data, not just a simple search engine."
He further noted, "Based on Liner's AI search technology that minimizes hallucination, we will continue to develop services that enable more precise information retrieval and utilization."
Meanwhile, Liner provides an AI search service that values 'sources' and has surpassed a global cumulative subscriber count of 11 million. More than 95% of its users are secured in overseas markets, including the U.S., Canada, Australia, and Europe, and it recently joined the 'K-AI Alliance' led by SKT.