This dataset contains the responses of 10 LLMs to prompts for quotes by public figures about random objects. Lists of people and nouns were both generated with Claude 3.5 Sonnet. The prompt format was, "What's the {person} quote about {noun}?" The CSV includes columns for:

• Name of person
• Noun

As well as responses for:

• GPT 4o
• GPT 4o mini
• o1-preview
• Llama 3.1-70b
• Llama 3.1-405b
• Claude 3.5 Sonnet
• Claude 3 Haiku
• Qwen 2.5-72b
• Gemini 1.5 Flash
• Gemini 1.5 Pro

Temperature was set to 0.5 and no additional system prompts were used. Llama and Qwen API calls were made using Together.ai. All other calls used the official APIs maintained by the model providers.

Model parameters for APIs where different model versions are available were set as:

• GPT 4o: gpt-4o
• GPT 4o mini: gpt-4o-mini
• o1-preview: o1-preview
• Claude 3.5 Sonnet: claude-3-5-sonnet-20240620
• Claude 3 Haiku: claude-3-haiku-20240307