Brave re-ran its own answer-quality benchmark in June 2026 and placed its chatbot first, ahead of Grok, Google AI Mode, ChatGPT and Perplexity. The company published the results on June 25 and promoted them again today through its company LinkedIn account. The scores exist to sell an API priced at $5 per 1,000 requests. In the first evaluation, conducted on November 30, 2025, Brave's chatbot finished second. The original post, titled Brave launches most powerful search API for AI to date, was published on February 12, 2026. Brave states that the June test repeated the method used in November, collecting answers from Ask Brave, Grok, Google AI Mode, ChatGPT and Perplexity across the same set of 1,500 queries randomly sampled from real-world usage. Judging was handled by large language models using both Claude Opus 4.5 and Claude Sonnet 4.5 in a majority vote setting. On a Bradley-Terry Elo rating, Ask Brave scored 1169 and Grok 1157, a gap of 12 points, while Google AI Mode sat at 1000, ChatGPT at 880 and Perplexity at 695.