Skip to content
TheChatArchiveThe Chat Archive
Geminigemini-2.5-pro-exp-03-25-grounding· GoogleHugging Face LMArena Search Arena 24K

the best llm on lmarena today? (only one)

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Geminigemini-2.5-pro-exp-03-25-grounding’s answer.
  1. User

    the best llm on lmarena today? (only one)

  2. Gemini Assistant

    Based on the latest available data from the Chatbot Arena Leaderboard, as of March 26, 2025, the top-ranked LLM is Gemini-2.5-Pro-Exp-03-25 from Google [8].

    Here's some additional context:

    • Source: The ranking comes from the LMSYS Chatbot Arena, a platform that uses crowdsourced, anonymous, randomized battles and an Elo rating system (now called Arena Score) to evaluate Large Language Models (LLMs) [1, 5, 6, 8][1].
    • Dynamic Rankings: Keep in mind that these rankings are dynamic and based on ongoing user votes (over 2.8 million as of late March 2025) [5, 8][2][3]. The leaderboard is updated frequently [1, 5][1][4].
    • Recent Top Models: Other models that have recently held top positions or are very close competitors include various versions of GPT-4 (like GPT-4o and GPT-4.5-Preview), Grok-3, and other Gemini models [8].
    • Multimodal vs. Text: There are separate leaderboards for text-only tasks and multimodal (text and image) tasks. While rankings are often similar, there can be differences [2]. The model mentioned above, Gemini-2.5-Pro-Exp-03-25, currently leads the overall leaderboard [8].
    • Open vs. Closed Source: The leaderboard includes both open-source and proprietary (closed-source) models[4]. While proprietary models often dominate the top spots, high-performing open-source models like DeepSeek-R1 have also recently ranked very highly [4, 8, 9].

    Therefore, based on the most recent data found, Gemini-2.5-Pro-Exp-03-25 is the single best-performing LLM on the Chatbot Arena leaderboard today, March 29, 2025 [8].

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.