Claude vs ChatGPT: which is better for writing tasks?

import anthropic
import os

client = anthropic.Anthropic(api_key=os.environ.get("ANTHROPIC_API_KEY"))

# Claude: system prompt is a first-class parameter, not part of messages
message = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=1024,
    system="You are a professional copywriter specializing in B2B SaaS. Write in active voice. Use exactly 3 sections. Include at least one specific benefit metric.",
    # Key difference: system is separate from messages, enabling precise control
    messages=[
        {
            "role": "user",
            "content": "Write a product description for an API rate-limiting service. Keep it under 300 words."
        }
    ]
)

print(message.content[0].text)

from openai import OpenAI
import os

client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))

# ChatGPT: system role must be in messages array as first role="system" entry
response = client.chat.completions.create(
    model="gpt-4o",
    max_tokens=1024,
    messages=[
        {
            "role": "system",
            "content": "You are a professional copywriter specializing in B2B SaaS. Write in active voice. Use exactly 3 sections. Include at least one specific benefit metric."
        },
        # Key difference: system is inside messages array, less explicit separation
        {
            "role": "user",
            "content": "Write a product description for an API rate-limiting service. Keep it under 300 words."
        }
    ]
)

print(response.choices[0].message.content)