Skip to content
TheChatArchiveThe Chat Archive
gpt-4o-search-preview-high-loc· OpenAIHugging Face LMArena Search Arena 24K

get a list of all the top LLMs and their benchmarks

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with gpt-4o-search-preview’s answer.
  1. User

    get a list of all the top LLMs and their benchmarks

  2. Assistant

    As of April 2025, several leading Large Language Models (LLMs) have demonstrated exceptional performance across various benchmarks. Below is a list of top LLMs along with their benchmark scores:

    Model NameMMLU ScoreGPQA ScoreHumanEval ScoreOrganizationRelease Date
    DeepSeek-R190.8%71.5%N/ADeepSeek2024
    Claude 3.5 Sonnet90.4%67.2%93.7%Anthropic2024
    o191.8%75.7%88.1%OpenAI2024
    GPT-4o88.0%53.6%N/AOpenAI2024
    Grok-287.5%56.0%88.4%xAI2024
    Llama 3.1 405B87.3%50.7%89.0%Meta2024
    Gemini 1.5 Pro75.8%N/AN/AGoogle2024

    Benchmark Descriptions:

    • MMLU (Massive Multitask Language Understanding): Evaluates a model's ability to handle diverse subjects, including science, humanities, and mathematics.

    • GPQA (Graduate-Level Professional Qualification Assessment): Assesses a model's performance on complex reasoning tasks at a graduate level.

    • HumanEval: Measures a model's proficiency in generating correct code solutions for programming problems.

    It's important to note that while these benchmarks provide valuable insights into model capabilities, recent discussions have highlighted concerns about benchmark contamination. Some models may have been inadvertently trained on benchmark datasets, potentially inflating their scores and not accurately reflecting their true generalization abilities[1].

    Therefore, while these scores offer a comparative overview, they should be interpreted with caution, considering the ongoing efforts to develop more robust and contamination-free evaluation methods.

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