zen4

Flagship

The primary Zen model. A 744B Mixture-of-Experts architecture with 40B active parameters per token, optimized for complex reasoning, multilingual understanding, and multi-domain tasks.

Specifications

Capabilities

• Complex multi-domain reasoning
• Long-form content generation
• Multilingual understanding (100+ languages)
• Instruction following and structured output
• 202K context window for large documents

API Usage

curl https://api.hanzo.ai/v1/chat/completions \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zen4",
    "messages": [{"role": "user", "content": "Hello, Zen."}]
  }'

from hanzoai import Hanzo

client = Hanzo(api_key="hk-your-api-key")

response = client.chat.completions.create(
    model="zen4",
    messages=[{"role": "user", "content": "Hello, Zen."}],
)
print(response.choices[0].message.content)

See Also

• zen4-ultra -- Maximum reasoning with chain-of-thought
• zen4-max -- Trillion-parameter frontier scale
• zen4-pro -- High capability at lower cost
• Pricing -- Full pricing table