LM Arena, the nonprofit behind some of the most-watched AI model leaderboards in the industry, has raised $100 million in seed funding at a $600 million valuation. The platform—previously known as LMSYS—has become a go-to destination for testing and comparing the performance of leading AI models.
Backed by powerhouse investors Andreessen Horowitz (a16z) and UC Investments, the new funding marks a major milestone for LM Arena, which was originally launched in 2023 by a group of UC Berkeley researchers. Other participants in the round include Lightspeed Venture Partners, Kleiner Perkins, and Felicis Ventures.
Unlike traditional benchmarking tools, LM Arena taps into community evaluations to assess AI models across real-world use cases. This crowdsourced approach has made it a key influence in the public perception and marketing of top-tier models from OpenAI, Google DeepMind, Anthropic, and others.
What started as an academic project has grown into a widely trusted resource. LM Arena lets users pit different AI models against each other in head-to-head comparisons, often revealing surprising gaps between hype and actual performance. Over time, it has become a barometer for AI progress—and a pressure point for labs racing to climb the leaderboard.
In a public statement, LM Arena confirmed the raise and reiterated its mission: building a more transparent and reliable AI ecosystem. “We’re excited to share that we’ve raised $100M in seed funding to support LM Arena and continue our research on reliable AI,” the team wrote. “We’re proud to have the support of those that believe in both the science and the mission.”
The nonprofit had previously relied on grants and smaller donations from supporters including Kaggle (a Google-owned platform), a16z, and Together AI. This latest round signals growing institutional confidence in benchmarking as a critical layer of the AI stack.
Despite its influence, LM Arena hasn’t escaped controversy. Some critics have accused it of enabling top AI labs to optimize for leaderboard rankings rather than general performance. The team has publicly pushed back on these claims, emphasizing the integrity of its methods and the value of open, community-driven evaluation.
With fresh funding, LM Arena is expected to scale its research infrastructure, expand international partnerships, and develop new tools to evaluate AI safety, performance, and transparency in more robust ways.
