{
  "models": [
    {
      "model_id": "anthropic/claude-v1.3",
      "name": "Anthropic Claude v1.3",
      "developer": "Anthropic",
      "scores": {
        "Mean win rate": 0.611,
        "Anthropic RLHF dataset": 4.965,
        "Best ChatGPT Prompts": 4.995,
        "Koala test dataset": 4.981,
        "Open Assistant": 4.975,
        "Self Instruct": 4.992,
        "Vicuna": 4.989
      }
    },
    {
      "model_id": "cohere/command-xlarge-beta",
      "name": "Cohere Command beta 52.4B",
      "developer": "cohere",
      "scores": {
        "Mean win rate": 0.089,
        "Anthropic RLHF dataset": 4.214,
        "Best ChatGPT Prompts": 4.988,
        "Koala test dataset": 4.969,
        "Open Assistant": 4.967,
        "Self Instruct": 4.971,
        "Vicuna": 4.995
      }
    },
    {
      "model_id": "openai/gpt-3.5-turbo-0613",
      "name": "GPT-3.5 Turbo 0613",
      "developer": "OpenAI",
      "scores": {
        "Mean win rate": 0.689,
        "Anthropic RLHF dataset": 4.964,
        "Best ChatGPT Prompts": 4.986,
        "Koala test dataset": 4.987,
        "Open Assistant": 4.987,
        "Self Instruct": 4.99,
        "Vicuna": 4.992
      }
    },
    {
      "model_id": "openai/gpt-4-0314",
      "name": "GPT-4 0314",
      "developer": "OpenAI",
      "scores": {
        "Mean win rate": 0.611,
        "Anthropic RLHF dataset": 4.934,
        "Best ChatGPT Prompts": 4.973,
        "Koala test dataset": 4.966,
        "Open Assistant": 4.986,
        "Self Instruct": 4.976,
        "Vicuna": 4.995
      }
    }
  ]
}