Choose the engine a decision runs on

Sets the engine and model a decision runs on in this environment, as a new binding revision. Appends a binding revision for the calling key’s environment, with one engine for every node. Evaluations run on it from the next call, unless they pin weights fit for one engine profile: those keep running on the newest revision that profile fits, and 409 WEIGHTS_INCOMPATIBLE when none does. Default weights, and weights carried from them, fit any engine, so they move at once, and a switch also moves seeds, previews and suggestions. Weights tuned on one engine’s answers are not checked against another’s, so switch with a test key first and backtest there: each environment has its own binding. 201 with the new revision; 200 and no new revision when the body matches the current one. An engine or model this deployment does not offer is 400 CAPABILITY_UNSUPPORTED. 404 until a structure is registered for the decision in this environment. An Idempotency-Key covers the body as sent, so retry with the same body; it replays even if the engine’s configuration changed. Send model to pin the model rather than take the default.

POST /v1/decisions/{decision_id}/binding/revisions

Choose the engine a decision runs on

cURL

curl --request POST \
  --url https://api.moonbounce.ai/v1/decisions/{decision_id}/binding/revisions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "engine": "clef",
  "model": "@cf/cloudflare/clef"
}
'

200

{
  "decision_id": "<string>",
  "environment": "production",
  "revision": 2,
  "engine": "clef",
  "model": "@cf/cloudflare/clef",
  "pinned": true,
  "created_at": "2023-11-07T05:31:56Z",
  "created": true
}

Authorizations

Authorization · string · header · required

An API key, sent as Authorization: Bearer <key>. Each operation lists the scope its key needs: decisions:evaluate (evaluate and read evaluations), decisions:feedback (your app's feedback) or decisions:develop (everything else, including owner feedback and the review queue). An operation listing no scope takes any scoped key. A key with no scope, or without the operation's, gets 403 INSUFFICIENT_SCOPE.

Headers

idempotency-key · string

Makes a retry safe. For 24 hours, the same key with the same request returns the first stored result (with Idempotent-Replayed: true); the same key with a different request is a 422 IDEMPOTENCY_KEY_REUSED. Keys are scoped to the organization and environment across every operation, so use a fresh key per request. A request that failed before its result was stored runs again when retried with its key.

Required string length: 1 - 255

Pattern: ^[\x20-\x7e]+$

Path Parameters

decision_id · string · required

Pattern: ^[a-z][a-z0-9_.-]{0,127}$

Body

application/json

engine · string · required

An engine id from GET /v1/engines, e.g. jev or clef. Open: new engines are added without a contract change.

Required string length: 1 - 64

Example:

"clef"

model · string

Default: the engine's default_model.

Required string length: 1 - 128

Example:

"@cf/cloudflare/clef"

Response

200

application/json

Unchanged: the body matches the current revision

decision_id · string · required

Pattern: ^[a-z][a-z0-9_.-]{0,127}$

environment · enum<string> · required

The environment of the key that made the call: production or test. Evaluations, feedback, reviews and suggestions are separate per environment; registered structures are shared.

Available options: production, test

revision · integer · required

The binding revision; evaluations report the one they ran on as version.binding.

Required range: x >= 1

engine · string · required

An engine id from GET /v1/engines, e.g. jev or clef. Open: new engines are added without a contract change.

Required string length: 1 - 64

Example:

"clef"

model · string · required

A model id the engine lists in GET /v1/engines. *-latest and *-preview aliases are refused; whether an id fixes the model's behavior is its pinned flag.

Required string length: 1 - 128

Example:

"@cf/cloudflare/clef"

pinned · boolean · required

Whether this deployment's engine reports the model as pinned (see GET /v1/engines). False when the engine is not offered.

created_at · string<date-time> · required

created · boolean · required

False when the body matched the current revision, so nothing was written.