⚡ Zen LM
Models

zen4-max

Trillion-parameter frontier MoE model. 1.04T (32B active) with 256K context.

zen4-max

Maximum Scale

Trillion-parameter frontier model with deep reasoning capabilities. 1.04T total parameters with 32B active per token -- the largest model in the Zen family.

Specifications

PropertyValue
Model IDzen4-max
Parameters1.04T (32B active)
ArchitectureMoE
Context Window256K tokens
Tierultra max
Input Price$3.60 / 1M tokens
Output Price$3.60 / 1M tokens

Capabilities

  • Frontier-scale reasoning with 1.04T parameters
  • Deep thinking with chain-of-thought capabilities
  • 256K context window for massive document processing
  • Strong performance on math, code, and complex reasoning
  • Efficient MoE inference with only 32B active parameters

API Usage

curl https://api.hanzo.ai/v1/chat/completions \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zen4-max",
    "messages": [{"role": "user", "content": "Summarize this research paper..."}]
  }'
from hanzoai import Hanzo

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

response = client.chat.completions.create(
    model="zen4-max",
    messages=[{"role": "user", "content": "Summarize this research paper..."}],
)
print(response.choices[0].message.content)

See Also

  • zen4 -- Flagship (744B MoE)
  • zen4-ultra -- Maximum reasoning with deep thinking
  • Pricing -- Full pricing table

On this page