Just eight months after launching its commercial service, AI leaderboard provider Arena, which originated as a research project at UC Berkeley in 2023, has reached $100 million in annualized run-rate revenue.

Arena is best known for its crowdsourced AI model performance leaderboard, generated from over 10 million user evaluations. Its consumer website lets users type a prompt and choose which of two models did the better job after receiving responses.

While Arena’s popular free leaderboard attracts community evaluators keen on early access to new AI models, it began generating revenue in September with its AI Evaluations service, providing model labs and enterprises with deep-dive performance analytics. The rapid growth indicates that businesses are as eager for these insights as the public is for model comparisons.

Despite not having direct competitors—Yupp’s shutdown left Arena unchallenged—the company competes ‘for the same dollar’ in post-training optimization services with human labeling startups like Mercor, Surge and Scale AI. With its recent $150 million Series A funding, its future looks bright.

Arena ranks models through various tasks including text, coding, vision and image generation, as well as complex workflows via Agent Mode. Co-founded by UC Berkeley postdoctoral students Anastasios Angelopoulos and Wei-Lin Chiang, and renowned professor Ion Stoica, Arena has raised a total of $250 million from prestigious investors.