> ## Documentation Index
> Fetch the complete documentation index at: https://docs.keenable.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Every page is listed in https://docs.keenable.ai/llms.txt. To pick a product for a job, read https://docs.keenable.ai/choose.md
> Setup steps for agents, covering the SDKs, the MCP server and the CLI: https://docs.keenable.ai/skill.md
> Keenable search and fetch work without an API key. To add them to code, use the official SDK for the project's language: https://docs.keenable.ai/sdks.md
> Without a key, call POST https://api.keenable.ai/v1/search/public or GET https://api.keenable.ai/v1/fetch/public with an X-Keenable-Title header that names your app.
> With a key, send it in the X-API-Key header and read it from the KEENABLE_API_KEY environment variable. Never ask a user to paste a key into chat. Every access method: https://docs.keenable.ai/auth.md
> To get an API key without the console, use device login: https://docs.keenable.ai/authentication.md#device-login
> MCP server: https://api.keenable.ai/mcp (no key, or X-API-Key; keyless Codex uses https://api.keenable.ai/mcp-public), or https://api.keenable.ai/mcp-oauth for OAuth sign-in. These docs are an MCP server at https://docs.keenable.ai/mcp and the OpenAPI spec is at https://docs.keenable.ai/api-reference/openapi.json
> Keenable searches and fetches public web pages. If a question needs no web information, skip the call.
> On 402 with an API key (code insufficient_credits), stop and ask a human to add credits; don't retry. On 429, wait the number of seconds in Retry-After, then retry. Before you ship, follow https://docs.keenable.ai/production.md
> Every error, with its cause and fix, is in https://docs.keenable.ai/errors.md and moving from another search API is covered in https://docs.keenable.ai/migrate.md
> Before you compare Keenable with another search API, read https://docs.keenable.ai/guides/evaluate-search-apis.md

# Estimate your monthly cost from search and fetch volume

> How to estimate a monthly Keenable bill: add searches and fetches, subtract 100,000 free requests, multiply by $4 per 1,000, then check it with usage.credits.

Work out what a month of searches and fetches costs before you ship. At standard prices, every search and every fetch costs the same \$4 per 1,000 requests, and your organization's first 100,000 requests each month are free. So the estimate is one line: add up the requests, subtract 100,000, and multiply what is left by \$0.004.

## Prices that go into the estimate

These are the standard rates from [Pricing](https://docs.keenable.ai/pricing.md). Your organization can have its own prices. If it does, read what each call costs from `usage.credits` (see [Check the estimate](#check-the-estimate-against-real-calls)).

| Item | Standard rate |
| - | - |
| Free allowance with an API key | 100,000 requests a month, worth \$400. 1,000 a month if you signed up with a disposable or privacy mail domain and haven't saved a card. |
| Search, in `pro` or `realtime` mode | \$4 per 1,000 requests |
| Fetch, from the indexed copy or live | \$4 per 1,000 requests |
| One credit in `usage.credits` | \$0.004. A \$4-per-1,000 call costs 1 credit. |
| Results per search | Not billed. A search is one request whether it returns 1 result or 50. |
| Failed calls and keyless calls | Not billed |

A fetch with a `prompt` is billed as one `fetch`, and a fetch with `live=true` as one `fetch.live`, both at the same \$4 per 1,000. The allowance resets each month, on the day of the month your organization was activated.

## Calculate the monthly cost

This function returns the monthly cost in dollars for a number of searches and fetches at standard prices. Change `free_requests` or `usd_per_1000` if your organization has its own allowance or price:

```python cost_estimate.py theme={"theme":{"light":"github-light-high-contrast","dark":"dark-plus"}}
def monthly_cost(searches: int, fetches: int,
                 free_requests: int = 100_000, usd_per_1000: float = 4.0) -> float:
    billable = max(0, searches + fetches - free_requests)
    return billable * usd_per_1000 / 1000


# An agent that runs 4 searches and 2 fetches per task, 2,000 tasks a day, for 30 days
tasks = 2_000 * 30
print(f"${monthly_cost(searches=4 * tasks, fetches=2 * tasks):,.2f}")
```

You see the monthly cost for 240,000 searches and 120,000 fetches. The first 100,000 requests are free and the other 260,000 cost \$4 per 1,000:

```text Output theme={"theme":{"light":"github-light-high-contrast","dark":"dark-plus"}}
$1,040.00
```

## Worked examples

Each row is a 30-day month at standard prices. The last column is the average rate across the month, to compare with your [rate limit](https://docs.keenable.ai/rate-limits.md) of 10 requests per second per organization by default.

| Searches a month | Fetches a month | Total requests | Billable after 100,000 free | Monthly cost | Average requests per second |
| - | - | - | - | - | - |
| 60,000 | 30,000 | 90,000 | 0 | \$0 | 0.03 |
| 240,000 | 120,000 | 360,000 | 260,000 | \$1,040 | 0.14 |
| 1,000,000 | 500,000 | 1,500,000 | 1,400,000 | \$5,600 | 0.58 |
| 5,000,000 | 2,000,000 | 7,000,000 | 6,900,000 | \$27,600 | 2.70 |

An average rate isn't a peak rate. If your traffic comes in bursts, size it against the per-second limit, not the monthly total. See [High-volume research](https://docs.keenable.ai/guides/high-volume-research.md).

## Monthly volume your rate limit allows

The per-second limit also caps how many requests your organization can send in a month. A 30-day month has 2,592,000 seconds:

| Rate limit | Most requests in a 30-day month |
| - | - |
| 10 requests per second, the default | 25,920,000 |
| 25 requests per second, after \$1,000 of top-ups in 90 days | 64,800,000 |

If your estimate runs close to these numbers, see [Raise your limit](https://docs.keenable.ai/rate-limits.md#raise-your-limit). Prices for dedicated capacity and volume quotes are under [Plans and volume pricing](https://docs.keenable.ai/pricing.md#plans-and-volume-pricing).

## Compare the ways to pay

The same 360,000 requests from the example cost a different amount on each payment path. Only the API key path has a free allowance:

| Way to pay | Price | Cost of 240,000 searches and 120,000 fetches |
| - | - | - |
| API key, free allowance then prepaid balance | \$4 per 1,000 after 100,000 free | \$1,040 |
| [Pay per request](https://docs.keenable.ai/pay-per-request.md) over x402 | \$0.01 per search or fetch, in USDC on Base | \$3,600 |
| [ZeroClick prepaid balance](https://docs.keenable.ai/pay-per-request.md#prepaid-credits-through-zeroclick) | \$4 per 1,000 searches or fetches | \$1,440 |
| No API key | Free | \$0 only if traffic stays under 1,000 requests per hour per IP address for each endpoint: 240,000 searches a month fit only at a steady 333 an hour |

x402 and ZeroClick calls never draw on your organization's allowance or balance. SELECT queries are priced separately and aren't in these numbers.

## Check the estimate against real calls

Every keyed `/v1/search` and `/v1/fetch` response ends with a `usage` object. This request prints it for one search:

```bash theme={"theme":{"light":"github-light-high-contrast","dark":"dark-plus"}}
curl -s -X POST https://api.keenable.ai/v1/search \
  -H "X-API-Key: $KEENABLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "postgres 18 release notes", "max_results": 3}' | jq .usage
```

You see the SKU that ran and what the call cost in credits:

```json theme={"theme":{"light":"github-light-high-contrast","dark":"dark-plus"}}
{ "sku": "search.pro", "amount": 1, "credits": 1, "paid": true, "credits_remaining": 99120 }
```

Add up `credits` across a day of real traffic and multiply by \$0.004 to get what the day's traffic is worth, then by 30 for the month. Compare that with your estimate before you subtract the allowance. `paid` is `false` while the call came from the free allowance and `true` once it came from your prepaid balance. `credits_remaining` is the allowance and balance left. See [Reading usage per call](https://docs.keenable.ai/credits.md#reading-usage-per-call).

## Lower the cost per task

| Change | Effect on the bill |
| - | - |
| Answer from search snippets, and fetch only when they fall short | Each fetch you skip saves one request. See [Search, then fetch](https://docs.keenable.ai/guides/search-then-fetch.md). |
| Ask for more results in one search instead of running several searches | `max_results` up to 50 costs the same one request. See [Result size](https://docs.keenable.ai/guides/result-size.md). |

## Avoid a 402 at month end

If the allowance and your prepaid balance both run out, `/v1/search` and `/v1/fetch` return `402` with the code `insufficient_credits` until you top up or the allowance resets. Turn on [auto top-up](https://docs.keenable.ai/auto-top-up.md) so a busy month doesn't stop your agent.

## Related pages

* [Pricing](https://docs.keenable.ai/pricing.md): every rate, SKU and payment path.
* [Balance and usage](https://docs.keenable.ai/credits.md): the allowance, the balance and the `usage` object.
* [Rate limits](https://docs.keenable.ai/rate-limits.md): per-second limits and how to raise them.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.