⚡ Zen LM
Models

zen4-ultra

Maximum reasoning model with 744B MoE (40B active) + extended chain-of-thought. 202K context.

zen4-ultra

Maximum Reasoning

The highest-capability Zen reasoning model. 744B MoE architecture with 40B active parameters and extended chain-of-thought for complex problems requiring deep multi-step reasoning.

Specifications

PropertyValue
Model IDzen4-ultra
Parameters744B (40B active)
ArchitectureMoE + CoT
Context Window202K tokens
Tierultra max
Input Price$3.00 / 1M tokens
Output Price$9.60 / 1M tokens

Capabilities

  • Extended chain-of-thought reasoning
  • Complex mathematical and scientific problem solving
  • Multi-step logical deduction
  • Research-grade analysis and synthesis
  • 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-ultra",
    "messages": [{"role": "user", "content": "Prove that the square root of 2 is irrational."}]
  }'
from hanzoai import Hanzo

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

response = client.chat.completions.create(
    model="zen4-ultra",
    messages=[{"role": "user", "content": "Prove that the square root of 2 is irrational."}],
)
print(response.choices[0].message.content)

See Also

  • zen4 -- Standard flagship without thinking mode
  • zen4-thinking -- Lightweight reasoning alternative
  • Pricing -- Full pricing table

On this page