CommencerCommencez gratuitement

Generating a product description

Imagine you're writing marketing copy for SonicPro headphones. Your goal is to generate a persuasive product description using DeepSeek.

Test how different prompting techniques, response lengths, and temperature settings influence the output. Since temperature only takes effect when reasoning is disabled, you'll disable reasoning via extra_body for this exercise.

Cet exercice fait partie du cours

<cours>Working with DeepSeek in Python</cours>
Voir le cours

Instructions de l’exercice

  • Create a detailed prompt to generate a product description for SonicPro headphones, including:
    • Active noise cancellation (ANC)
    • 40-hour battery life
    • Foldable design
  • Disable reasoning by passing extra_body={"thinking": {"type": "disabled"}}.
  • Experiment with max_tokens and temperature to see how they affect the output.

Exercice interactif pratique

Essayez cet exercice en complétant ce code d’exemple.

client = OpenAI(api_key="", base_url="https://api.together.xyz/v1")

# Create a detailed prompt
prompt = """
____
"""

response = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4-Pro",
    # Disable reasoning so temperature takes effect
    extra_body=____,
    messages=[{"role": "user", "content": prompt}],
    # Experiment with max_tokens and temperature settings
    max_tokens=____,
    temperature=____
)

print(response.choices[0].message.content)
Modifier et exécuter le code