Diffbot review: the short version
Diffbot is a ai scraper from Diffbot, and in our 2026 assessment it scores 8/10 overall. It is at its best for enterprise entity data and zero-maintenance extraction, it starts at from $299 / mo, and it expects rotating residential, plus an unlocker for protected sources behind it. Per-credit enterprise pricing; the knowledge graph is licensed separately from extraction APIs.
The rest of this review covers how it actually works, what it costs at realistic volumes, which proxy type to pair it with, how it behaves against anti-bot systems, where it breaks, and which alternatives make more sense for adjacent workloads. Every figure below reflects list pricing and hands-on testing rather than vendor marketing copy.
- +Automatic page-type classification
- +Article, product, discussion and image APIs
- +Knowledge Graph with billions of entities
- +Crawlbot for whole-domain jobs
What Diffbot is and how it works
## Diffbot review: the short version
Diffbot is a ai scraper from Diffbot, and in our 2026 assessment it scores 8/10 overall. It is at its best for enterprise entity data and zero-maintenance extraction, it starts at from $299 / mo, and it expects rotating residential, plus an unlocker for protected sources behind it. Per-credit enterprise pricing; the knowledge graph is licensed separately from extraction APIs.
The rest of this review covers how it actually works, what it costs at realistic volumes, which proxy type to pair it with, how it behaves against anti-bot systems, where it breaks, and which alternatives make more sense for adjacent workloads. Every figure below reflects list pricing and hands-on testing rather than vendor marketing copy.
- Automatic page-type classification
- Article, product, discussion and image APIs
- Knowledge Graph with billions of entities
- Crawlbot for whole-domain jobs
## What Diffbot is and how it works
Diffbot approaches extraction as a vision and classification problem rather than a parsing problem. Its models look at the rendered page, decide what kind of page it is, and return a normalised object — article with author and date, product with price and SKU, organisation with employees and funding — using the same schema whether the source is a Fortune 500 site or a regional blog.
That uniformity is the product. For media monitoring, competitive intelligence and firmographic enrichment, where the source list is unbounded and constantly changing, no selector-based approach can keep up. The Knowledge Graph goes a step further by linking extracted entities into a queryable graph, which is data you cannot generate by crawling alone.
Price positions it firmly in enterprise. At $299 monthly entry and no visibility into proxies or retries, it is the wrong tool for a team scraping twenty known competitor sites — Scrapy plus residential proxies does that for a fraction of the cost. It is the right tool when the alternative is hiring people to maintain parsers for ten thousand domains.
## Diffbot scorecard and measured performance
Scores below are relative to the other tools in this directory, not to software in general — a 6 for scale still means a tool that handles more traffic than most projects will ever generate. The performance figures come from crawling a mixed basket of static HTML, JavaScript-rendered commerce and lightly protected listing pages from three regions.
Read them alongside your own target list. The tool almost never determines success rate on its own; the combination of exit IP quality, request fingerprint and request pacing does, which is why two teams running the same framework routinely report success rates thirty points apart.
## Throughput, rendering and resource profile
Throughput numbers only mean something with the cost attached. Diffbot delivers Managed, bulk crawls supported, and can render JavaScript, which is convenient and roughly five to twenty times more expensive per page than a plain fetch.
Use these figures to size infrastructure before committing to a plan or a proxy contract. Work backwards from records per day, apply a realistic success rate, add a retry factor of 1.2–1.6, and only then choose concurrency.
## Diffbot pricing and real cost per thousand pages
Per-credit enterprise pricing; the knowledge graph is licensed separately from extraction APIs.
Two bills run in parallel: bandwidth and tokens. Bandwidth behaves like any crawl, but token cost scales with page size, so pruning the DOM before it reaches the model is the highest-leverage optimisation available — boilerplate removal alone can cut input tokens by 70%. Use a cheap small model for extraction and reserve frontier models for reasoning over the extracted data, not for reading raw HTML.
A useful discipline: express every option as cost per thousand usable records, not cost per month. A plan that looks cheap and delivers a 60% success rate is more expensive than a premium option at 95%, because the failures consume bandwidth, retries, engineering attention and calendar time.
## Best proxies for scraping with Diffbot
AI scrapers still fetch pages over the network, so they inherit every proxy question a classic crawler has. The difference is corpus breadth: retrieval pipelines pull from hundreds of unrelated domains, which makes a rotating residential gateway with country targeting the sane default rather than a per-site decision. Keep politeness high — a knowledge base built by hammering a small publisher is a reputational problem as much as a technical one.
With Diffbot specifically, identity attaches through fully managed — no proxy configuration exposed. Get that wiring right before tuning anything else — a rotation bug that reuses one exit across a thousand requests will look exactly like a bad proxy provider.
## How Diffbot handles anti-bot systems
Most RAG corpora come from documentation, blogs, forums and public data where bot management is light, and a well-behaved crawler with residential exits is enough. The exceptions are commerce and social sources, which are hardened precisely because their data is valuable. Route those through an unlocker rather than escalating your own fingerprint work, and honour robots.txt and licensing — content provenance is now an audit question in any organisation shipping AI features.
Practically, treat detection as a budget rather than a binary. Measure success rate per domain daily, escalate a domain one tier at a time — better headers, then better IPs, then a browser, then an unlocker — and stop at the first tier that clears your threshold. Escalating everything to the most expensive tier is the most common and most costly mistake in scraping operations.
## Scaling Diffbot in production
Scale AI ingestion by shrinking the input, not by adding workers. Map the site first and crawl only the URLs your index needs, deduplicate near-identical pages by content hash, chunk at semantic boundaries, and cache raw fetches so re-embedding never re-crawls. Store the source URL, fetch timestamp and licence with every chunk; retrieval quality and legal defensibility both depend on that metadata.
- Track success rate, cost per thousand records and bytes per page as your three primary metrics
- Retire proxy sessions automatically on repeated failures instead of retrying blindly
- Deduplicate URLs before dispatch — duplicates cost bandwidth, credits and rate-limit headroom
- Validate content, not just HTTP status: a 200 that returns a consent wall is a failed fetch
- Keep a second fetching path warm so a vendor incident degrades throughput instead of stopping it
## Diffbot pros and cons
No scraping tool is universally correct; each one trades cost, control and maintenance in a different ratio. Diffbot makes the following trade explicitly.
## Who should use Diffbot — and who should not
Choose Diffbot when your workload looks like enterprise entity data and zero-maintenance extraction and your team is comfortable with HTTP API · Crawlbot · KG query language. It fits organisations that have already decided whether they are buying outcomes or building capability, because it sits clearly on one side of that line: you buy an outcome and trade unit cost for removed maintenance.
Look elsewhere if you need deterministic extraction of financial or pricing fields at high volume, where selector-based parsing remains cheaper and auditable.
## Compliance and responsible collection
Collecting publicly accessible data is broadly lawful in most jurisdictions, but the surrounding obligations are real: respect robots.txt where it expresses the publisher's intent, avoid authentication walls you have not been granted access to, never collect personal data without a lawful basis under GDPR or equivalent, and keep request rates low enough that you never degrade the target's service.
Reputable proxy providers enforce KYC precisely because misuse of their networks is their liability as well as yours. Document what you collect, why, how long you retain it and who can access it. For AI training corpora, record licensing and provenance per source — that record is increasingly the first thing an auditor or enterprise customer asks to see.
Diffbot scorecard and measured performance
Scores below are relative to the other tools in this directory, not to software in general — a 6 for scale still means a tool that handles more traffic than most projects will ever generate. The performance figures come from crawling a mixed basket of static HTML, JavaScript-rendered commerce and lightly protected listing pages from three regions.
Read them alongside your own target list. The tool almost never determines success rate on its own; the combination of exit IP quality, request fingerprint and request pacing does, which is why two teams running the same framework routinely report success rates thirty points apart.
| Criterion | Score | Assessment |
|---|---|---|
| Ease of adoption | 10/10 | Productive on day one |
| Scale ceiling | 9/10 | Comfortable at millions of pages |
| Anti-bot resilience | 8/10 | Good with the right proxies |
| Documentation | 8/10 | Adequate; community fills the gaps |
| Value for money | 5/10 | Premium pricing — justify it with volume |
Throughput, rendering and resource profile
Throughput numbers only mean something with the cost attached. Diffbot delivers Managed, bulk crawls supported, and can render JavaScript, which is convenient and roughly five to twenty times more expensive per page than a plain fetch.
Use these figures to size infrastructure before committing to a plan or a proxy contract. Work backwards from records per day, apply a realistic success rate, add a retry factor of 1.2–1.6, and only then choose concurrency.
| Metric | Observed | Notes |
|---|---|---|
| Throughput | Managed, bulk crawls supported | Per worker or per plan tier, on a stable target |
| JavaScript rendering | Yes, vision-based analysis | Rendering multiplies cost 5–20× versus plain HTTP |
| Memory footprint | None (remote) | Sizing input for container limits |
| Success profile | Very high on articles, products and organisation pages | Depends far more on proxy quality than on the tool |
| Proxy support | Fully managed — no proxy configuration exposed | How identity is attached to a request |
Diffbot pricing and real cost per thousand pages
Per-credit enterprise pricing; the knowledge graph is licensed separately from extraction APIs.
Two bills run in parallel: bandwidth and tokens. Bandwidth behaves like any crawl, but token cost scales with page size, so pruning the DOM before it reaches the model is the highest-leverage optimisation available — boilerplate removal alone can cut input tokens by 70%. Use a cheap small model for extraction and reserve frontier models for reasoning over the extracted data, not for reading raw HTML.
A useful discipline: express every option as cost per thousand usable records, not cost per month. A plan that looks cheap and delivers a 60% success rate is more expensive than a premium option at 95%, because the failures consume bandwidth, retries, engineering attention and calendar time.
| Plan | Price | What you get |
|---|---|---|
| Free trial | $0 | 10k credits, 14 days |
| Startup | $299 / mo | 250k credits, all extraction APIs |
| Plus | $899 / mo | 1M credits, knowledge graph access |
| Enterprise | custom | Bulk crawls, SLA, custom entities |
Best proxies for scraping with Diffbot
AI scrapers still fetch pages over the network, so they inherit every proxy question a classic crawler has. The difference is corpus breadth: retrieval pipelines pull from hundreds of unrelated domains, which makes a rotating residential gateway with country targeting the sane default rather than a per-site decision. Keep politeness high — a knowledge base built by hammering a small publisher is a reputational problem as much as a technical one.
With Diffbot specifically, identity attaches through fully managed — no proxy configuration exposed. Get that wiring right before tuning anything else — a rotation bug that reuses one exit across a thousand requests will look exactly like a bad proxy provider.
| Target profile | Proxy type | Typical price | Why |
|---|---|---|---|
| Internal APIs, open data, docs sites | Datacenter | $0.30 – $2.00 / IP / mo | No consumer-IP requirement; cheapest possible bandwidth |
| Mid-tier commerce, listings, forums | Rotating residential | $1.00 – $8.00 / GB | Real ISP-assigned IPs clear reputation checks |
| Logged-in accounts, dashboards | ISP / static residential | $1.50 – $6.00 / IP / mo | One stable identity per account, held for months |
| App-only endpoints, hardest anti-bot | Mobile (4G/5G) | $4.00 – $20.00 / GB | Carrier CGNAT makes per-IP blocking costly for the target |
| Everything already blocked | Unlocker API | $0.50 – $3.00 / 1k requests | Challenge solving handled provider-side, billed per success |
Article extraction and a Crawlbot job
# 1) Single page — automatic type detection
curl "https://api.diffbot.com/v3/analyze?token=TOKEN&url=https%3A%2F%2Fexample.com%2Fnews%2F1"
# 2) Whole-domain crawl with automatic extraction
curl "https://api.diffbot.com/v3/crawl" \
-d token=TOKEN \
-d name=competitor-news \
-d seeds=https://competitor.example.com \
-d apiUrl=$(python -c "import urllib.parse;print(urllib.parse.quote('https://api.diffbot.com/v3/article'))") \
-d maxToCrawl=50000
# 3) Collect results
curl "https://api.diffbot.com/v3/crawl/data?token=TOKEN&name=competitor-news"How Diffbot handles anti-bot systems
Most RAG corpora come from documentation, blogs, forums and public data where bot management is light, and a well-behaved crawler with residential exits is enough. The exceptions are commerce and social sources, which are hardened precisely because their data is valuable. Route those through an unlocker rather than escalating your own fingerprint work, and honour robots.txt and licensing — content provenance is now an audit question in any organisation shipping AI features.
Practically, treat detection as a budget rather than a binary. Measure success rate per domain daily, escalate a domain one tier at a time — better headers, then better IPs, then a browser, then an unlocker — and stop at the first tier that clears your threshold. Escalating everything to the most expensive tier is the most common and most costly mistake in scraping operations.
Scaling Diffbot in production
Scale AI ingestion by shrinking the input, not by adding workers. Map the site first and crawl only the URLs your index needs, deduplicate near-identical pages by content hash, chunk at semantic boundaries, and cache raw fetches so re-embedding never re-crawls. Store the source URL, fetch timestamp and licence with every chunk; retrieval quality and legal defensibility both depend on that metadata.
- +Track success rate, cost per thousand records and bytes per page as your three primary metrics
- +Retire proxy sessions automatically on repeated failures instead of retrying blindly
- +Deduplicate URLs before dispatch — duplicates cost bandwidth, credits and rate-limit headroom
- +Validate content, not just HTTP status: a 200 that returns a consent wall is a failed fetch
- +Keep a second fetching path warm so a vendor incident degrades throughput instead of stopping it
Diffbot pros and cons
No scraping tool is universally correct; each one trades cost, control and maintenance in a different ratio. Diffbot makes the following trade explicitly.
| Strengths | Weaknesses |
|---|---|
| Genuinely zero selector maintenance | The most expensive option in this directory |
| Consistent schema across millions of unrelated sites | No proxy-level control at all |
| Knowledge graph adds firmographic context no crawler produces | Overkill for a handful of known targets |
| Mature, stable API since 2012 | Custom fields need enterprise engagement |
Who should use Diffbot — and who should not
Choose Diffbot when your workload looks like enterprise entity data and zero-maintenance extraction and your team is comfortable with HTTP API · Crawlbot · KG query language. It fits organisations that have already decided whether they are buying outcomes or building capability, because it sits clearly on one side of that line: you buy an outcome and trade unit cost for removed maintenance.
Look elsewhere if you need deterministic extraction of financial or pricing fields at high volume, where selector-based parsing remains cheaper and auditable.
Compliance and responsible collection
Collecting publicly accessible data is broadly lawful in most jurisdictions, but the surrounding obligations are real: respect robots.txt where it expresses the publisher's intent, avoid authentication walls you have not been granted access to, never collect personal data without a lawful basis under GDPR or equivalent, and keep request rates low enough that you never degrade the target's service.
Reputable proxy providers enforce KYC precisely because misuse of their networks is their liability as well as yours. Document what you collect, why, how long you retain it and who can access it. For AI training corpora, record licensing and provenance per source — that record is increasingly the first thing an auditor or enterprise customer asks to see.
Diffbot FAQs
How much does Diffbot cost?+
Paid plans start around $299 per month for 250k credits, with knowledge graph access on higher tiers.
Do I need my own proxies with Diffbot?+
No. Proxying, rendering and retries are entirely managed and not exposed.
When is Diffbot the wrong choice?+
When you scrape a small, known set of sites — deterministic selectors plus your own proxies are dramatically cheaper.
Keywords covered
ai scraping proxy · scraping proxy providers · entity extraction api · scraping proxies comparison
