⚡ Zen LM
Models

zen-agent

32B dense model with tool use and planning for agentic AI workflows.

zen-agent

Agentic AI

A 32B dense transformer designed for agentic workflows with native tool use, multi-step planning, and autonomous execution. Excels at breaking down complex tasks, calling external tools, and iterating on results.

This model is in preview. Access may be limited.

Specifications

PropertyValue
Model IDzen-agent
Parameters32B
ArchitectureDense
Context Window131K tokens
StatusPreview
HuggingFace--

Capabilities

  • Native tool/function calling
  • Multi-step task planning and execution
  • Code execution and iteration
  • Web browsing and information retrieval
  • File manipulation and data processing
  • Self-correction and error recovery
  • 131K context for complex agent workflows

Usage

API

from hanzoai import Hanzo

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

tools = [
    {
        "type": "function",
        "function": {
            "name": "search_web",
            "description": "Search the web for information",
            "parameters": {
                "type": "object",
                "properties": {
                    "query": {"type": "string", "description": "Search query"},
                },
                "required": ["query"],
            },
        },
    },
]

response = client.chat.completions.create(
    model="zen-agent",
    messages=[{"role": "user", "content": "Find the latest research on quantum error correction and summarize the key findings."}],
    tools=tools,
    tool_choice="auto",
)

message = response.choices[0].message
if message.tool_calls:
    for call in message.tool_calls:
        print(f"Tool: {call.function.name}")
        print(f"Args: {call.function.arguments}")
else:
    print(message.content)

See Also

  • zen4 -- 744B MoE flagship model
  • zen-coder -- 32B code-specialized model
  • zen-pro -- 32B professional foundation model

On this page