Skip to content

Correct the AIMultiple benchmark date, scope, and tier framing #1363

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
hmishra2250 merged 3 commits into from
Sep 8, 2026
+14 −2
Merged

Correct the AIMultiple benchmark date, scope, and tier framing #1363

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
641b615
docs(search,extract): add measured-performance table and credit-seman…
hmishra2250 Sep 4, 2026
28c6a31
docs(search,extract): fix measured-performance date, tier framing, an…
hmishra2250 Sep 4, 2026
3c66fc3
docs(search): note the AIMultiple benchmark queries are the source's …
hmishra2250 Sep 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions features/extract.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ import ExtractWithoutURLsCURL from "/snippets/v2/extract/without-urls/curl.mdx";

The `/extract` endpoint simplifies collecting structured data from any number of URLs or entire domains. Provide a list of URLs, optionally with wildcards (e.g., `example.com/*`), and a prompt or schema describing the information you want. Firecrawl handles the details of crawling, parsing, and collating large or small datasets.

<Info>We've simplified billing so that Extract now uses credits, just like all of the other endpoints. Each credit is worth 15 tokens.</Info>
<Info>We've simplified billing so that Extract now uses credits, just like all of the other endpoints. Each credit is worth 15 tokens. See [Billing](/billing) for how credits map to plan pricing; scrape and crawl are billed at 1 credit per page ([Scrape cost](/features/scrape#scraping-a-url-with-firecrawl)).</Info>

## Using `/extract`

Expand Down Expand Up @@ -218,7 +218,7 @@ curl -X POST https://api.firecrawl.dev/v2/extract \

## Billing and Usage Tracking

We've simplified billing so that Extract now uses credits, just like all of the other endpoints. Each credit is worth 15 tokens.
We've simplified billing so that Extract now uses credits, just like all of the other endpoints. Each credit is worth 15 tokens. See [Billing](/billing) for how credits map to plan pricing; scrape and crawl are billed at 1 credit per page ([Scrape cost](/features/scrape#scraping-a-url-with-firecrawl)).

You can monitor Extract usage via the [dashboard](https://www.firecrawl.dev/app/extract).

Expand Down
12 changes: 12 additions & 0 deletions features/search.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -611,6 +611,18 @@ For more details about the scraping options, refer to the [Scrape Feature docume
> Are you an AI agent that needs a Firecrawl API key? See [firecrawl.dev/agent-onboarding/SKILL.md](https://www.firecrawl.dev/agent-onboarding/SKILL.md) for automated onboarding instructions.


## Measured Performance

Independent evaluators have benchmarked Firecrawl Search against other search APIs. The figures below are pulled directly from each evaluator's published results as of the date shown.

| Metric | Value | Source | Date | n / ± |
| --- | --- | --- | --- | --- |
| Agent Score (Mean Relevant × Quality) | 14.58, rank 2 of 8 (Mean Relevant 4.30/5, Quality 3.39/5) | [AIMultiple — Agentic Search Benchmark](https://aimultiple.com/agentic-search) | Dec 2025 (source updated May 25, 2026) | 95% CI 13.12–15.98, n=100 AI/LLM-domain queries drawn from AIMultiple's own organic search traffic (10,000 bootstrap resamples); the source notes results don't generalize to other domains |

<Note>
These are third-party measurements, each with its own harness, task set, and methodology — they are not directly comparable to one another or to Firecrawl's own benchmarks. Point-estimate rank 2 of 8; paired-bootstrap testing found no statistically significant gap versus Brave, Exa, or Parallel Search Pro. Firecrawl's own first-party benchmark pages, including the open-source Developer Index evaluation, live at [firecrawl.dev/benchmarks](https://www.firecrawl.dev/benchmarks).
</Note>

## Search feedback

When a search result is useful or misses important content, submit feedback with `POST /v2/search/{jobId}/feedback`. The first feedback submission for a search job can refund 1 credit, subject to team limits, and helps improve Firecrawl search quality. See [Search Feedback](/api-reference/endpoint/search-feedback).