Proxy Tiers & Credit Rules
The Reader API supports four proxy tiers. Start with the default (free). Escalate to Shared Pool, Datacenter, or Residential only when needed — each step costs more credits but unlocks harder-to-access content.
Try in Playground →Proxy Tier Selector
param: proxyReader API base cost is 2 credits per request. Each proxy tier adds credits on top.
Direct — no proxy
Standard outbound request with no proxy. Covers the vast majority of public web pages.
2 base + 0 = 2 cr
When: Start here. Works for most news sites, blogs, documentation, and product pages.
Shared pool of datacenter IPs
Routes through a shared pool of datacenter IPs. Bypasses basic anti-bot filters.
2 base + 2 = 4 cr
When: Use when Standard Mode returns blocked or empty content on lightly protected sites.
Dedicated datacenter IPs
Dedicated datacenter IPs with stronger reputation. Handles moderate anti-bot measures.
2 base + 5 = 7 cr
When: E-commerce sites, job boards, and pages with moderate bot protection.
⚠️ Replaces old proxy:1 "Bypass Mode" (was 5 cr). New proxy:1 is now the lighter shared-pool tier.
Real residential ISP IPs
Real residential ISP IPs. Highest success rate against aggressive bot detection.
2 base + 10 = 12 cr
When: Ticketing platforms, social networks, paywalled content, and heavily rate-limited sites.
Best practice: Always start at proxy: 0.
Only step up if the response returns empty markdown or an error.
This keeps credit costs minimal.
// POST /api/v1/url
{
"t": "url",
"s": "https://target-site.com/page",
"proxy": 0, // 0 = direct (free)
// 1 = shared pool (+2 cr)
// 2 = datacenter (+5 cr)
// 3 = residential (+10 cr)
"d": 20000
} Code Examples
# Direct (default) — most sites
curl -X POST "https://www.searchcans.com/api/v1/url" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"t":"url","s":"https://example.com","proxy":0}'
# Shared Pool — lightly protected sites
curl -X POST "https://www.searchcans.com/api/v1/url" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"t":"url","s":"https://example.com","proxy":1}'
# Datacenter — moderate protection
curl -X POST "https://www.searchcans.com/api/v1/url" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"t":"url","s":"https://example.com","proxy":2}'
# Residential — heavy protection
curl -X POST "https://www.searchcans.com/api/v1/url" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"t":"url","s":"https://example.com","proxy":3}' Credit Cost per Request
Credits are only deducted on successful responses (code: 0).
Blocked pages, timeouts, and errors are not charged — even if a higher proxy tier was used.
Reader API base cost is 2 credits. Each Bypass Mode tier adds credits on top. Full breakdown: Pricing page.
| proxy | Mode (as shown on Pricing page) | Base | Add-on | Total / Request |
|---|---|---|---|---|
| 0 | Standard Mode | 2 cr | — | 2 cr |
| 1 | Bypass Mode (Proxy:1) | 2 cr | +2 cr | 4 cr |
| 2 | Bypass Mode (Proxy:2 #Datacenter) | 2 cr | +5 cr | 7 cr |
| 3 | Bypass Mode (Proxy:3 #Residential) | 2 cr | +10 cr | 12 cr |
What Affects Your Credit Cost?
Deducts Credits
- • Successful response only — credits deducted when
code: 0is returned - • proxy tier add-on — +2 / +5 / +10 credits on top of the 2 cr base (see table above)
- • Batch pages (SERP only) —
pageparameter: +1 credit per additional page beyond the first
No Extra Credits
- • Failed requests — blocked pages, timeouts, and errors (
code ≠ 0) are never charged - • file: 1 — parse PDF / DOCX / XLSX. Included in the Reader API base cost, no add-on
- • image: 1 / 2 — screenshot first screen or full page. Included in the Reader API base cost, no add-on
- • mode, html, w, d, b — all behavior/render parameters carry no additional credit cost
Escalation Strategy
Start Low
Always start with proxy: 0. Most public pages don't need any proxy. This keeps your credit cost at the baseline 2 credits per request.
Step Up on Failure
If you receive empty markdown, a blocked response, or repeated errors, increment the tier by 1 and retry. Stop as soon as you get a successful result.
Cache the Result
Once you find the minimum working tier for a given domain, store it. On subsequent requests to the same domain, skip directly to the known-good tier to save both time and credits.
Anti-Scraping Limits
Some sites deploy advanced bot detection that no proxy can defeat (e.g. Cloudflare v3 challenges, behavioral fingerprinting). Even proxy: 3 cannot guarantee 100% success on these sites.
⚠️ Migration Note — proxy: 1 Behavior Change
In earlier API versions, proxy: 1 activated "Bypass Mode" at 5 credits/request. That behavior has been remapped to proxy: 2 (Datacenter, +5 credits). If your integration used proxy: 1 for high-success-rate extraction, update it to proxy: 2.
Site Type → Recommended Tier
| Site Type | Examples | Recommended proxy |
|---|---|---|
| News / Blogs / Docs | Reuters, GitHub, Wikipedia, Medium | 0 |
| Marketing / SaaS Sites | Stripe, Notion, Product Hunt | 0 |
| E-commerce (basic) | Shopify stores, small retailers | 1 |
| Job Boards | Indeed, Glassdoor, LinkedIn jobs | 1–2 |
| E-commerce (strict) | Amazon, Etsy, eBay product pages | 2 |
| Social Platforms | Reddit, Twitter/X, Facebook | 2–3 |
| Ticketing / Finance | Ticketmaster, Bloomberg, Seeking Alpha | 3 |
| Paywalled Content | WSJ, FT, academic journals | 3 |
Ready to extract any page?
Get 100 free credits on sign up. No credit card required.