# FRED · Series observations

Get market and financial data with FRED · Series observations from Alexandria through one Spiral API key. The only capability in this provider that returns numbers; everything else exists to find the id it takes. Two independent time axes: `observation_start`/`observation_end` bound when the economy did the thing, `realtime_start`/`realtime_end` bound when FRED knew it. It runs in the background and returns structured JSON. Pay as you go, with no subscription.

- Provider: Alexandria
- Category: Finance
- Runs: In the background
- Returns: json
- Tool id: `alexandria` / `fred/series/observations`
- Page: https://spiral.ai/tools/alexandria/fred/series/observations

## Pricing

$0.001746 per call.

## Limits

- Longest queue wait: 1 hour
- Run deadline: 5 minutes
- Longest run, end to end: 225 minutes
- Largest inline input: 1 MB
- Largest input file: 2 GB
- Largest output: 5 GB

## Input

| Field | Kind | Allowed | Default | Description |
| --- | --- | --- | --- | --- |
| `series_id` (required) | text |  | — | The series to read, for example GNPCA or UNRATE. Find one with `series/search`. |
| `limit` | number | … to 100000 | 100000 | Observations to return. |
| `units` | one of | lin, chg, ch1, pch, pc1, pca, cch, cca, log | lin | Value transformation: lin levels, chg change, ch1 change from a year ago, pch percent change, pc1 percent change from a year ago, pca compounded annual rate of change, cch continuously compounded rate of change, cca continuously compounded annual rate of change, log natural log. |
| `offset` | number |  | 0 | Observations to skip. |
| `frequency` | one of | d, w, bw, m, q, sa, a, wef, weth, wew, wetu, wem, wesu, wesa, bwew, bwem | — | Aggregate down to a lower frequency. Asking for a frequency higher than the series' native one is an error rather than a no-op, and asking for the native one does nothing. Omit for no aggregation. |
| `sort_order` | one of | asc, desc | asc | Sort by observation date. |
| `output_type` | one of | 1, 2, 3, 4 | 1 | 1 observations by real-time period, 2 all observations by vintage date, 3 new and revised observations only, 4 initial release only. The fields described below are the shape of output_type 1; 2 and 3 replace the record entirely with one column per vintage date. |
| `realtime_end` | text |  | — | End of the real-time period, YYYY-MM-DD. Defaults to today. |
| `vintage_dates` | text |  | — | Comma-separated YYYY-MM-DD dates to read the data as it stood on, instead of a real-time period. Capped server-side by output type - roughly 2000 dates for json, far fewer for output_type 2 on a daily series. |
| `realtime_start` | text |  | — | Start of the real-time period, YYYY-MM-DD. Defaults to today, meaning the current vintage. `vintage_dates` is an alternative to this pair, not a filter layered on top of it. |
| `observation_end` | text |  | 9999-12-31 | Latest observation date to return, YYYY-MM-DD. |
| `observation_start` | text |  | 1776-07-04 | Earliest observation date to return, YYYY-MM-DD. When the economy did the thing - not to be confused with `realtime_start`, which is when FRED knew it. |
| `aggregation_method` | one of | avg, sum, eop | avg | How `frequency` aggregates. No effect unless `frequency` is set. |

## Example

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "alexandria",
    "model": "fred/series/observations",
    "input": {
      "series_id": "<series_id>"
    }
  }'
```

Try it in the Playground: https://spiral.ai/start?provider=alexandria&model=fred%2Fseries%2Fobservations

## Common questions

### FRED · Series observations

**How much does FRED · Series observations cost?**

$0.001746 per call. You pay as you go from one Spiral balance, with no subscription, and a failed run is not charged.

**How do I use FRED · Series observations through an API?**

Send `POST /v1/runs` with `"provider": "alexandria"`, `"model": "fred/series/observations"` and the input fields on this page, using one Spiral API key. It runs in the background: poll the run, or name a webhook to be told when it finishes.

**Do I need an account with Alexandria to use FRED · Series observations?**

No. Spiral runs it on its own Alexandria account and charges your Spiral balance.

**Can my AI agent use FRED · Series observations?**

Yes. Agents find and run it through Spiral's MCP server, CLI or agent skill, with the same API key.

