---
title: "Read a durable managed model billing receipt"
method: GET
path: "/v1/managed-models/requests/{id}"
tags: ["Managed models"]
---

# Read a durable managed model billing receipt

`GET /v1/managed-models/requests/{id}`

## Request

- Base URL: `/` (relative, and the document was published on a file host, not at the API’s origin)
- URL: `/v1/managed-models/requests/{id}`
- Auth: HTTP bearer

Returns a receipt owned by the authenticated account. Running or unknown requests keep their spending reservation until reconciled. The receipt contains usage and charge evidence without the prompt or generated answer. It is not a response replay endpoint.

## Path parameters

- `id` string, required

## Response `200`

Durable request accounting state.

- ManagedModelReceipt: Durable billing evidence scoped to the authenticated account. Prompts, model answers, credentials and private pricing are not included.
  - `id` string, required
  - `model` string, required: Model ID returned by this account’s managed model catalog.
  - `price_version` string, required
  - `state` 'running' | 'succeeded' | 'failed' | 'unknown', required: Running and unknown keep their reservation. Succeeded posts the verified charge. Failed records a definitive rejection without a model charge.
  - `reserved_micros` integer, required: Original spending reservation in USD micros. It is held only while state is running or unknown.
  - `charge_micros` integer, required: Settled customer charge in USD micros. Zero until successful settlement. Model charges are separate from compute and saved storage.
  - `usage` union, required: Verified normalized usage after successful settlement. Empty for running, unknown and failed requests.
    - ManagedModelUsage: Normalized, verified accounting counters. These receipt counters differ from the native usage object returned by a completion request.
      - `input_tokens` integer, required: Total input including all cache buckets. Uncached input is the remainder after subtracting the cache buckets.
      - `output_tokens` integer, required: Total output including reasoning tokens.
      - `cached_input_tokens` integer, required: Input tokens read from cache.
      - `cache_creation_5m_input_tokens` integer, required: Input tokens written with a five-minute cache lifetime.
      - `cache_creation_1h_input_tokens` integer, required: Input tokens written with a one-hour cache lifetime.
    - object
  - `created_at` string, date-time, required
  - `finished_at` string, date-time: Present for succeeded and failed requests.

## Other responses

- `404`: The receipt does not exist in this account.
- `default`: Authentication or accounting access failed.

## Changes

- **2026-10-01** `ba84c74aeec8`: 1 info
  - endpoint added
- **2026-10-01** `544fef290a54`: 1 breaking
  - api path removed without deprecation
- **2026-09-27** `c27e8c7c4e1f`: 1 info
  - endpoint added
- **2026-09-27** `90c4cb8514af`: 1 breaking
  - api path removed without deprecation
- **2026-09-25** `808adc26fb17`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/nodus-compute/apis/nodus-infrastructure-execution-api/changes/v1/managed-models/requests/:id/get.md)

---

[API](https://skmtc.dev/nodus-compute/apis/nodus-infrastructure-execution-api.md) - [All operations](https://skmtc.dev/nodus-compute/apis/nodus-infrastructure-execution-api/llms.txt) - [OpenAPI document](https://skmtc.dev/nodus-compute/apis/nodus-infrastructure-execution-api/revisions/a59210899d95?raw)