Skip to content
Hanzo AI
Try Hanzo
Language: English (United States)

Zen ModelsComplete API Catalog

Zen 6 and Zen 6 Flash available now · Zen 7 in research preview

Built for agentic coding and multimodal work. 19 foundation models from Zoo Labs Foundation across language, code, vision, image, audio, and retrieval. Zen MoDE (Mixture of Diverse Experts) architecture. From $0.01/MTok.

Get API keyExplore ModelsFull Pricing

Quick Start

Drop-in OpenAI-compatible API. Switch models with one line.

python
from hanzoai import Hanzo

client = Hanzo()

# Use any Zen model
response = client.chat.completions.create(
    model="zen6",           # or zen6-flash, zen5-coder, etc.
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
Models
19
Max Context
2M
From
$0.01/MTok
Architectures
MoE + Dense
Modalities
7

Zen5 -- Agentic Frontier

Zen5

Agentic frontier models trained on 10B+ tokens of real-world tool use and multi-step reasoning.

ultra maxpro max
ModelParamstotal params active per token contextContextArchAvailability
  1. Zen5 (Default)35B (3B active)256KZen frontier MoE$0 / $0

    Canonical Zen5 default. a 35B frontier MoE (3B active) (35B total / 3B active per token, released Apr 2026, Apache-2.0). 256K context, agentic-trained, OpenAI + Anthropic API. The everyday Zen5 chat model.

  2. Zen5 Coder80B (MoE)256KZen Coder MoE$0 / $0

    Code-specialized Zen5 tier. 80B sparse MoE tuned for repo-scale code understanding, agentic refactoring, and tool-use coding loops.

  3. Zen5 Flash4B32KZen dense$0 / $0

    Smallest and cheapest text-only Zen5 chat tier. 4B-class dense, sub-100ms TTFT, 32K context. For high-volume routing and simple agent loops.

  4. Zen5 Mini230B (10B active)192KZen MoE$0 / $0

    Frontier agentic at the lowest cost in the family. Built on a 230B agentic MoE (10B active) (230B MoE / 10B active, released Feb 2026). 80.2% SWE-Bench Verified, 76.3% BrowseComp; trained on 200K+ real-world environments via large-scale RL.

  5. Zen5 Pro284B (37B active)1MZen Flash MoE$0 / $0

    Zen Flash IQ2_XXS-imatrix weights (81 GB GGUF on zenlm/zen-5-pro-gguf). 284B total / 37B active per token, 1M context, asymmetric routed-MoE quant. Fits a single 128 GB Apple Silicon / DGX Spark / H100 80 GB.

Ready to build with Zen?

All 19 Zen models available via OpenAI-compatible API. Start building today.

Get StartedTry Hanzo DevView Pricing