Skip to content
TheChatArchiveThe Chat Archive
DeepSeekdeepseek-v3-0324· DeepSeekHugging Face LMArena Human Preference 140K

That was a nice try but it is a person that ate a bird, I want you to make a bird that…

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Claudeclaude-opus-4-20250514-thinking-16k’s answer.
  1. User

    That was a nice try but it is a person that ate a bird, I want you to make a bird that eats a bird!

  2. DeepSeek Assistant

    Got it! Here’s a more literal (and dramatic) ASCII art of a bird eating another bird—this time with a raptor (eagle/hawk) gripping a smaller bird in its talons:

                              _,._      
                            .'   _`.    
                           /   _`   \   
                          |   (o)    |  
                       .-.\    `    /.-.
                      (   `'.      .'   )
                       '-._  `'--'`  _.-'
                           `"------"`
                             / |  | \
                            /  |  |  \
                           /   |  |   \
                          /    |  |    \
                         /     |  |     \
                        /      |  |      \
                       '-------'  '-------'
                      (A larger bird swooping)
                             \|  |/
                              \  /
                               ||
                               ||
                            .-'  `-.
                           /        \
                          |   o  o   |
                           \   __   /
                            `-.  .-'
                               \/
    

    Key Details:

    • Top: The larger bird (raptor) with outstretched wings mid-flight.
    • Middle: Its talons (||) gripping the smaller bird.
    • Bottom: The prey bird (simplified, with o for eyes).

    Want it even bigger or more detailed? Let me know!

    (Note: ASCII art has limits—this is abstract but captures the idea!)

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