Create model

POST /v1/risk/state/model

Publishes your organisation's model as a NAMED VALUE, so a decision taken today can be reconstructed tomorrow and a change made today can be undone.

It answers with a NAME and not with the state. The masses stay on your organisation's own encrypted store and are referred to by an address computed from their own content: the shape, the geometry seed, the position in the window, the threshold, the masses themselves as IEEE-754 bits, and the fold watermark behind them. That is what makes the value nameable without making the caller its custodian.

IT IS IDEMPOTENT ON THE VALUE. A model that has not changed publishes to the name it already has and mints nothing, reporting minted=false — so publishing at every boundary that matters is free. Ten values are retained per organisation, bounded in BYTES rather than in rows, and the oldest is disposed of past that.

A model that has learned nothing is refused: planted is not learned, and a value that reproduces nothing is not a value.

It is POST and PUT on one address because they are one plane's two verbs over one kind of thing: POST mints a value from the model in force, PUT puts a value in force. They were /v1/risk/state/snapshot and /v1/risk/state/restore — two addresses named after the operation rather than after the thing, which is how a reader ends up asking what the difference between a snapshot and a value is.

Request

The document declares no body for POST /v1/risk/state/model. The handler is typed in cloud but its shape is not yet emitted, so the fields are not listed here — ask MCP's describe for riskPublishModel, which answers from the running route.

Response

201 body — 9 fields.

Failure carries the platform error shape — see Errors.

Examples

hanzo risk state model create

import { Configuration, RiskApi } from 'hanzoai';

const api = new RiskApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.riskPublishModel();

from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import RiskApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = RiskApi(client).risk_publish_model()

cfg := hanzoai.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := hanzoai.NewAPIClient(cfg)

resp, _, err := client.RiskAPI.RiskPublishModel(context.Background()).Execute()
if err != nil {
	return err
}

use hanzo_client::apis::{configuration::Configuration, risk_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = risk_api::risk_publish_model(&cfg, Default::default()).await?;

import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.RiskApi;

ApiClient client = new ApiClient();
client.setBearerToken(System.getenv("HANZO_API_KEY"));

var result = new RiskApi(client).riskPublishModel();

curl -X POST https://api.hanzo.ai/v1/risk/state/model \
  -H "Authorization: Bearer $HANZO_API_KEY"

curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "risk",
         "arguments": {
           "op": "riskPublishModel",
           "input": {}
         }
       }
     }'

Risk API · All Hanzo APIs · Interactive reference