mirror of
https://github.com/hanzo-apps/hanzo.ai.git
synced 2026-10-30 14:58:08 +00:00
zen-image, zen-voice, zen-music, zen-video and zen-rerank have no account behind them: /v1/models lists none, and a call answers 404 or 500. llms.txt offered them, the models page featured zen-image and the lineup counted zen-rerank as served.
68 lines
2.6 KiB
Plaintext
68 lines
2.6 KiB
Plaintext
# Hanzo
|
|
|
|
> Hanzo OS is the operating system for agentic companies.
|
|
|
|
Hanzo OS unifies Enso, AI coworkers, secure execution, company data, full-stack
|
|
observability and the cloud beneath them in one open system. Four layers —
|
|
company, intelligence, execution, infrastructure — with learning and governance
|
|
cutting across all of them.
|
|
|
|
Run Hanzo OS on Hanzo Cloud or your own infrastructure.
|
|
|
|
Enso is the intelligence layer: one endpoint that reads the objective, applies
|
|
your policies and puts the right model behind the work. Zen is the open-weight
|
|
family. Both run beside 500+ models on the same endpoint and the same key.
|
|
|
|
Kai is the decision model: typed questions about a state (choice, noul, score),
|
|
answered with calibrated distributions instead of generated text, at
|
|
`POST https://api.hanzo.ai/v1/decisions`. Docs: https://docs.hanzo.ai/docs/decisions
|
|
|
|
Hanzo AI Inc, Techstars '17.
|
|
|
|
Install: `curl -fsSL hanzo.sh | sh`
|
|
API reference for agents: https://docs.hanzo.ai/llms.txt
|
|
Skills for agents: https://hanzoskills.com/llms.txt
|
|
|
|
## About
|
|
Hanzo AI (Techstars '17) trains its own models and operates the infrastructure
|
|
that serves them. Two model families are ours — enso and Zen — and they run on
|
|
Hanzo Cloud alongside the rest of the platform: inference, agents, embeddings,
|
|
vector search, GPU compute, storage, and identity, on one host behind one key.
|
|
|
|
## Models
|
|
- enso — our frontier family. Three tiers: `enso-flash`, `enso`, `enso-ultra`.
|
|
- Zen — our open-weight family. `zen5`, `zen5-pro`, `zen5-coder`, `zen5-flash`
|
|
for text and code; `zen-vl` for vision; `zen-embedding` for retrieval;
|
|
`zen-guard` for safety classification. Weights are published.
|
|
- Models from other labs are reachable on the same endpoint, with the same key,
|
|
when a workflow needs a specific one.
|
|
|
|
## Products
|
|
- AI API — chat, embeddings, images, and reranking at api.hanzo.ai/v1
|
|
- Hanzo Dev — coding agent
|
|
- AI Studio — train, fine-tune, and deploy models
|
|
- Hanzo Platform — managed AI infrastructure with GPU clusters
|
|
- Model Context Protocol (MCP) — open protocol for AI context management
|
|
- Agent frameworks for building autonomous systems
|
|
|
|
## Links
|
|
- [Website](https://hanzo.ai)
|
|
- [Documentation](https://docs.hanzo.ai)
|
|
- [Models](https://hanzo.ai/models)
|
|
- [AI Studio](https://studio.hanzo.ai)
|
|
- [Hanzo Platform](https://platform.hanzo.ai)
|
|
- [Hanzo Dev](https://hanzo.ai/dev)
|
|
- [Zen models](https://zenlm.org)
|
|
- [GitHub](https://github.com/hanzoai)
|
|
- [Status](https://status.hanzo.ai)
|
|
|
|
## API
|
|
- Base URL: https://api.hanzo.ai/v1
|
|
- Authentication: API key via Bearer token
|
|
- Docs: https://docs.hanzo.ai/docs/api
|
|
|
|
## Contact
|
|
- Support: [email protected]
|
|
- Sales: [email protected]
|
|
- X: @hanzoai
|