Diffbot Review
Turns any web page into structured data using automatic AI extraction, plus a pre-built knowledge graph of billions of facts
Is this your product? Claim this page · Request a change
Looking for a Diffbot alternative? See our ranked comparison.→What is Diffbot?
Diffbot is an API that reads a web page and figures out what's on it: article text, product details, discussion threads, and so on, without you writing per-site parsing rules. Point it at a URL and it returns clean structured JSON, using computer vision and machine learning trained on billions of pages rather than site-specific selectors.
On top of the extraction APIs, Diffbot runs its own Knowledge Graph, a pre-built database of entities (people, companies, products) that it has already crawled and structured from the public web. You can query that graph directly using DQL (Diffbot Query Language) instead of scraping and parsing pages yourself, which is the part that sets it apart from most scraping tools.
Everything is metered on a single credit system: a basic page extract costs 1 credit, while exporting a Knowledge Graph entity record costs 25 credits and a facet query costs 100, so budgeting takes some getting used to. Site-wide crawling (Crawlbot) and Bulk Extract are also gated to the Plus plan and above; the Free and Startup plans only cover single-page Extract, Natural Language, and Knowledge Graph Search/Enhance.
Diffbot screenshots
Who it's for
- ✓ Teams that want structured article, product, or discussion data without maintaining custom scrapers per site
- ✓ Developers who want to query a pre-built knowledge graph of companies and people instead of crawling and parsing that data themselves
- ✓ Ops and research teams comfortable working through an API or Google Sheets integration rather than a point-and-click scraper builder
Who should look elsewhere
- ✗ Small projects with light, occasional scraping needs, since the entry paid plan is a flat $299/month
- ✗ Teams that need site-wide crawling on a budget, since Crawl and Bulk Extract only unlock at the $899/month Plus plan
- ✗ Teams that want a visual, no-code scraper builder rather than an API-first workflow
- ✗ Anyone who needs precise, predictable per-page cost, since credit consumption varies a lot by endpoint
Pros
- + Automatic extraction works across many page types (articles, products, discussions) without per-site rules
- + The Knowledge Graph gives you access to pre-structured entity data you'd otherwise have to build yourself
- + Free tier has no credit card requirement and includes Extract, Natural Language, and Knowledge Graph access at lower volume
- + No long-term contract required on any plan; Diffbot says you can cancel any time and won't be billed for subsequent months
Cons
- – Credit consumption differs sharply by feature: a basic page extract costs 1 credit, but a Knowledge Graph entity export costs 25 credits and a facet query costs 100, so the headline credit counts are misleading until you've used it
- – Crawl and Bulk Extract aren't available on Free or Startup; you need the $899/month Plus plan to get site-wide crawling
- – Entry paid tier jumps straight to $299/month with no smaller paid step between Free and Startup
- – API-first workflow and its own query language (DQL) have a real learning curve for teams used to point-and-click scraping tools
- – Overages on paid plans are billed pro rata automatically, so a usage spike can show up as an unexpected charge
Diffbot pricing
At about $299/month to start, it sits at the higher end of Web Scraping & Data Extraction pricing.
| Plan | Price | Highlights |
|---|---|---|
| Free | Free | 10,000 credits/month · 5 calls per minute · Extract, Natural Language, and Knowledge Graph Search/Enhance (Bulk Extract and Crawl require Plus or higher) · No credit card required |
| Startup | $299/mo | 250,000 credits/month ($0.001 per credit) · 5 calls per second · Multiple user licenses · Same product access as Free; Bulk Extract and Crawl are not included |
| Plus | $899/mo | 1,000,000 credits/month ($0.0009 per credit) · 25 calls per second · Adds Bulk Extract and 25 active crawls · 3 user licenses |
| Enterprise | Custom | Custom credit allotment and rate · 100+ active crawls · 25+ calls per second · Custom SLA, managed solutions, and a shared Slack channel |
Everything runs on a shared credit system, and the cost per action varies a lot: a basic page extract is 1 credit (2 with a datacenter proxy), a Knowledge Graph entity export is 25 credits, and a Knowledge Graph facet query is 100 credits, so the headline credit count on a plan doesn't map cleanly to a fixed number of pages. Crawl and Bulk Extract are gated to the Plus plan and above; Free and Startup only include Extract, Natural Language, and Knowledge Graph Search/Enhance. Overage on paid plans is billed pro rata at your plan's per-credit rate; free-plan usage past 10,000 credits or the rate limit returns a 429 error instead. Startup and Plus are flat monthly fees with no annual discount shown on the pricing page. Diffbot also gives free Startup-tier access to students and academic researchers.
Pricing verified July 17, 2026 · source
How Diffbot's pricing compares
Diffbot next to its closest alternatives on entry price, billing, and whether pricing is public.
| Tool | Starting price | Billing | Free option | Pricing disclosed |
|---|---|---|---|---|
| Diffbot | $299/mo | usage-based | Yes | Partly public |
| Apify | $29/mo | tiered | Yes | Partly public |
| Bright Data | Usage-based | usage-based | Yes | Partly public |
| Zyte | Usage-based | usage-based | Yes | Partly public |
| ScrapingBee | $49/mo | usage-based | Trial (1,000 API credits, no credit card required) | Partly public |
| Firecrawl | $19/mo | tiered | Yes | Partly public |
| Octoparse | $69/mo | tiered | Yes | Partly public |
Is Diffbot still actively developed?
Last significant update: June 2026. Improved date sorting (sortBy:date) in Knowledge Graph article search, letting you filter between the last six months and the full archive.
Top Diffbot alternatives
Diffbot FAQ
Is Diffbot free?+
Yes, there's a permanent free plan with 10,000 credits a month and no credit card required, covering Extract, Natural Language, and Knowledge Graph Search/Enhance at low volume. Crawl and Bulk Extract aren't included until the Plus plan. No free trial is needed since the free tier itself doesn't expire.
How much does Diffbot cost for a small team?+
The first paid plan, Startup, is $299/month for 250,000 credits and 5 calls per second. It doesn't add Crawl or Bulk Extract access; those require the $899/month Plus plan. There's no smaller paid tier between the free plan and Startup.
Why do Diffbot credits run out faster than expected?+
Credits are shared across very different operations and priced very differently: a basic page extract is 1 credit, but exporting a Knowledge Graph entity record is 25 credits and a Knowledge Graph facet query is 100 credits. Heavy Knowledge Graph use can burn through a plan's monthly allotment much faster than simple page extraction.
Does Diffbot offer custom pricing?+
Yes. Enterprise plans are quote-only and cover custom credit allotments, 100+ active crawls, and a custom SLA with managed solutions.