Zen LM by Hanzo AI -- frontier models for code, reasoning, vision, multimodal, embeddings, and safety

Zen LM

Zen LM by Hanzo AI is a frontier model family spanning code, reasoning, vision, multimodal, embeddings, and safety. Over 1T parameters. From the 4B zen3-nano for edge deployment to 1T+ parameters for frontier-scale reasoning, Zen MoE (Mixture of Experts) delivers state-of-the-art performance across every major AI task.

Model Lineup

Zen4 Generation

9 models -- flagship, reasoning, and code. From 8B zen4-mini to ~400B zen4-ultra.

Zen3 Generation

5 models -- multimodal, vision, edge, safety, and embeddings.

API Reference

Every Zen model at api.hanzo.ai, behind one key.

Pricing

Transparent pricing from $0.30/MTok.

Quick Start

Zen models are served at https://api.hanzo.ai/v1. Get a key from the console, then:

curl https://api.hanzo.ai/v1/chat/completions \
  -H "Authorization: Bearer hk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{"model": "zen5", "messages": [{"role": "user", "content": "Hello, Zen."}]}'

Python

from hanzoai import Hanzo

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

response = client.chat.completions.create(
    model="zen5",
    messages=[{"role": "user", "content": "Hello, who are you?"}],
)
print(response.choices[0].message.content)

TypeScript

import Hanzo from '@hanzo/ai'

const client = new Hanzo({ apiKey: 'hk-your-api-key' })

const response = await client.chat.completions.create({
  model: 'zen5',
  messages: [{ role: 'user', content: 'Hello, who are you?' }],
})
console.log(response.choices[0].message.content)

Key Facts

Zen models covering all major AI tasks
Zen MoDE (Mixture of Diverse Experts) architecture
Apache 2.0 weights -- download and run the same models yourself
Transparent pricing from $0.30/MTok

Wire compatibility

/v1/chat/completions and /v1/embeddings take and return the standard chat-completions JSON shape, so an HTTP client already written against that shape works here once its base URL points at https://api.hanzo.ai/v1 and it sends a Hanzo key:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.hanzo.ai/v1",
    api_key="hk-your-api-key",
)

response = client.chat.completions.create(
    model="zen5",
    messages=[{"role": "user", "content": "Hello, who are you?"}],
)

Model ids do not carry across vendors — zen5 is a Hanzo model and resolves only here.

API key prefix: hk-
Endpoint: api.hanzo.ai/v1

All Models

Loading pricing...

Introduction