AI模型评测

  - name: MMLU
    desc: 大规模多任务语言理解基准 
    avatar: /img/mmlu-benchmark-icon.png
    link: https://paperswithcode.com/sota/multi-task-language-understanding-on-mmlu 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: Open LLM Leaderboard
    desc: Hugging Face推出的开源大模型排行榜单 
    avatar: /img/huggingface-icon.png
    link: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: C-Eval
    desc: 一个全面的中文基础模型评估套件 
    avatar: /img/c-eval-icon.png
    link: https://cevalbenchmark.com/static/leaderboard_zh.html 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: FlagEval
    desc: 智源研究院推出的FlagEval（天秤）大模型评测平台 
    avatar: /img/flageval-icon.png
    link: https://flageval.baai.ac.cn/#/trending 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: SuperCLUE
    desc: 中文通用大模型综合性测评基准 
    avatar: /img/superclue-icon.png
    link: https://www.cluebenchmarks.com/static/superclue.html 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: OpenCompass
    desc: 上海人工智能实验室推出的大模型开放评测体系 
    avatar: /img/opencompass-icon.png
    link: https://opencompass.org.cn/leaderboard-llm 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: CMMLU
    desc: 一个综合性的大模型中文评估基准 
    avatar: /img/cmmlu-benchmark-icon.png
    link: https://github.com/haonan-li/CMMLU 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: MMBench
    desc: 全方位的多模态大模型能力评测体系 
    avatar: /img/mmbench-icon.png
    link: https://mmbench.opencompass.org.cn/leaderboard 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: HELM
    desc: 斯坦福大学推出的大模型评测体系 
    avatar: /img/helm-benchmark.png
    link: https://crfm.stanford.edu/helm/latest 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: Chatbot Arena
    desc: 以众包方式进行匿名随机对战的LLM基准平台 
    avatar: /img/lmsys-icon.png
    link: https://chat.lmsys.org 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: LLMEval3
    desc: 由复旦大学NLP实验室推出的大模型评测基准 
    avatar: /img/llm-eval-icon.png
    link: http://llmeval.com/index 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: H2O EvalGPT
    desc: H2O.ai推出的基于Elo评级方法的大模型评估系统 
    avatar: /img/h2o-evalgpt-icon.png
    link: https://evalgpt.ai 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'
  - name: PubMedQA
    desc: 生物医学研究问答数据集和模型得分排行榜 
    avatar: /img/pubmedqa-icon.png
    link: https://pubmedqa.github.io 
    bgColor: '#FCDBA0'
    textColor: '#A05F2C'