Models

After this page you know which family answers which kind of question, which ones you can download, and which ids the gateway serves right now.

Five roles

The split is by the shape of the answer. An open question — write, explain, plan — needs generation, so it goes to Zen. A bounded one — which model, which tool, is this done, allow or deny — has a known answer space, so it goes to Kai, which answers it without generating text. Creative media synthesis — video, images, 3D, audio — runs on proprietary Creative AI checkpoints in Studio. Enso sits in front and sends each operation to the one that should answer it. Policy has the last word on any action, and a model can make a verdict stricter, never looser.

Other lines

The live catalogue

Every model below answers on one address with one key. Pass the id as model and the rest of the request is the same whichever you pick. The one exception is kai, listed with "outputs": ["decision"]: it answers typed questions on POST /v1/decisions, not chat.

curl https://api.hanzo.ai/v1/chat/completions \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"<id from below>","messages":[{"role":"user","content":"Hello"}]}'

The table is fetched from /v1/models in your browser when this page loads, so it is the catalogue the gateway is serving right now rather than a copy written down when the page was built. The same list, by family, needs no key:

curl -s https://api.hanzo.ai/v1/models \
  | jq -r '.data[] | select(.id | test("^(zen|enso)")) | "\(.id)\t\(.owned_by)\t\(.pricing.input_per_million)/\(.pricing.output_per_million)"'

What a rate means

Rates are per million tokens (pricing.input_per_million, output_per_million; pricing.prompt and completion state the same rates per token, as OpenRouter does), input and output priced separately, and they are the same whether you call from the CLI, an SDK, HTTP or an MCP tool — the price is a property of the model, not of the surface you reach it through.

Pricing explains how usage becomes a bill. Credits covers what happens when the balance runs out.

Hugging Face Hub & Model Registry

Hanzo AI maintains a public model and system card registry across Hugging Face (huggingface.co/hanzoai) and ZenLM (huggingface.co/zenlm), organizing its first-party architecture across 5 functional families:

1. Kai — Unified Multimodal Decision Intelligence (hanzoai/kai-*): Calibrated, typed decision models. Hosted API on POST /v1/decisions (kai-1.3, kai-1.4, kai-1.5, kai-1.6, and the unified multimodal target kai-1.7), alongside specialized baselines (kai-1-agent, kai-1-multilingual).
2. Enso — Intelligent Routing (hanzoai/enso-*): Dynamic prompt routing across cost, latency, and quality tiers (enso-auto, enso-flash, enso-pro, enso-ultra, enso-free).
3. Zen — Generative Intelligence (zenlm/zen6*): Open-weight foundation models for agentic coding, reasoning, and multimodality (zen6, zen6-flash, zen6-coder, zen-vl, zen-embedding, zen-reranker).
4. Guard — Safety & Oversight (hanzoai/guard, zooai/lifeguard): Deterministic policy gateways and consequence-grounded evaluation suites (redirected from zoo-labs/lifeguard).
5. Agent Systems (hanzoai/kata, hanzoai/gen): Autonomous DAG execution and software engineering agent runtimes.

Interactive Spaces

Try models and inspect benchmarks in real-time on Hugging Face Spaces:

• Leaderboard: Open Decision Model Arena comparing Kai, Jev 1.13, and frontier LLMs.
• Kai Playground: Interactive typed question console (boolean, choice, score).
• Enso Router Simulator: Real-time prompt complexity and cost router preview.
• Guard Arena: Adversarial prompt injection and policy boundary testing.

Specifications