SearchCans: AI Data Infrastructure — SERP API + Reader API
SearchCans is the data infrastructure layer for AI applications. Use the SERP API to search Google and Bing, then the Reader API to convert any URL into clean, LLM-ready Markdown. Both APIs use POST requests with JSON payloads and Bearer token authentication. This Python example demonstrates the complete dual-engine workflow: search for "AI Agents" via SERP API, extract the first result URL, and convert it to structured Markdown via Reader API — the standard pipeline for AI agents, RAG systems, and LLM applications.
SearchCans provides enterprise-grade data infrastructure at $0.56 per 1,000 requests. Parallel Lanes let you run more requests at the same time, and eligible paid plans can add lanes together with Lane Stacking. Start with prepaid credit packs and use one pool of credits across SERP API and Reader API.
AI Data Infrastructure The Data Layer
Your AI Needs
Search Google and Bing, turn any URL into clean LLM-ready Markdown, and run more requests at the same time with Parallel Lanes. Add more eligible plans later if you need more throughput. Starting at $0.56/1K.
New users get 100 free credits upon registration.
1 Parallel Lane = 1 request running now. More lanes mean more work can run at once.
import requests
# 1. Search Google (SERP API)
search_resp = requests.post(
"https://www.searchcans.com/api/search",
headers="Authorization": "Bearer YOUR_KEY",
json=
"s": "AI Agents", # Search query
"t": "google", # Engine type
"d": 20000 # 20s server timeout
,
timeout=30 # 30s client timeout
)
# Get the first URL result
first_url = search_resp.json()['data'][0]['url']
# 2. Extract Content (Reader API)
reader_resp = requests.post(
"https://www.searchcans.com/api/url",
headers="Authorization": "Bearer YOUR_KEY",
json=
"s": first_url, # Source URL
"t": "url",
"b": True, # Use browser rendering
"w": 3000 # Wait 3s for JS (use 5000ms+ for heavy sites)
,
timeout=30 # 30s client timeout
)
# Output clean LLM-ready Markdown
print(reader_resp.json()['data']['markdown'])Parallel Lanes explained
Understanding Parallel Lanes
Parallel Lanes show how much work your account can do at the same time. Your speed depends on how many requests can run now, not on an hourly reset.
1 lane = 1 live request
If you have 3 lanes, 3 requests can run at the same time.
No waiting for an hourly reset
When one request finishes, that lane is ready again right away.
Built for agents and batch jobs
More lanes help with bursts, queues, and busy workflows that send many requests at once.
Simple speed estimate
Requests/minute ≈ Lanes × (60 ÷ 2.5s) This quick estimate uses a typical ~2.5s average request time for standard SERP requests. Reader API and heavier pages can take longer.
1 lane ≈ 24 requests/minute
3 lanes ≈ 72 requests/minute
22 lanes ≈ 528 requests/minute
Need more throughput later?
Start with the lanes you need today. If your jobs start to queue later, add another eligible plan and your lanes add up in one account.
AI-ready facts
Quick buying facts for lanes, credits, and throughput
These are the three facts most buyers need before they compare plans.
- What is a Parallel Lane?
- A Parallel Lane is one request that can run right now. When one request finishes, that lane is ready again for the next request.
- How are credits used?
- Standard Google and Bing search use 1 credit per request. Reader API standard mode uses 2 credits per request. More lanes do not change the credit cost of one request.
- Why do lanes matter?
- They help agents, automations, and batch jobs finish faster when many requests come in at the same time.
Two Powerful APIs, One Platform
SERP API + Reader API: The complete data infrastructure for AI applications
Add Lanes as You Grow
Start with the lanes you need now. Add another eligible plan later when your workload grows.
URL to Markdown API
Turn messy web pages into clean Markdown that is ready for RAG, agents, and data pipelines.
Lightning Fast
Average latency of 2–3 seconds per request. Optimized for AI agent workloads.
99.99% Uptime
Enterprise-grade reliability with redundant infrastructure. Your AI applications run without interruption.
Built for Developers
Get structured JSON from search and content APIs that fit cleanly into LangChain, LlamaIndex, and custom workflows.
Ready to integrate? Check out our comprehensive documentation
AI-Grade Search Infrastructure
Start with the lanes and credits you need today. Add another eligible plan later if your workload grows.
All plans include full access to SERP API and Reader API
Standard
- Get 20,000 credits
- 2 Parallel Lanes ? What are Parallel Lanes? A lane is one request running now. This plan has 2 lanes, so 2 requests can run at the same time. When one finishes, that lane is ready again.
- Real-time results
- Fast response time
- Email technical support
- No Lane Stacking
Credits valid for 6 months
Starter
- Get 132,000 credits
- 3 Parallel Lanes ? What are Parallel Lanes? A lane is one request running now. This plan has 3 lanes, so 3 requests can run at the same time. When one finishes, that lane is ready again.
- Real-time results
- Faster response priority
- Email technical support
- Eligible for Lane Stacking ? What is Lane Stacking? Add another eligible plan and your lanes add up in one account. More lanes mean more requests can run at the same time. 1× $99 = 3 Lanes 2× $99 = 6 Lanes $99 + $597 = 25 Lanes
Credits valid for 6 months
Pro
- ~1 Million credits
- 22 Parallel Lanes ? What are Parallel Lanes? A lane is one request running now. This plan has 22 lanes, so 22 requests can run at the same time. When one finishes, that lane is ready again.
- Real-time results
- Priority Request Routing
- Tier-1 Developer Support
- Eligible for Lane Stacking ? What is Lane Stacking? Add another eligible plan and your lanes add up in one account. More lanes mean more requests can run at the same time. 1× $99 = 3 Lanes 2× $99 = 6 Lanes $99 + $597 = 25 Lanes
Credits valid for 6 months
Ultimate
- Get 3,000,000 credits
- 68 Parallel Lanes ? What are Parallel Lanes? A lane is one request running now. This plan has 68 lanes, so 68 requests can run at the same time. When one finishes, that lane is ready again.
- Dedicated Cluster Node
- Zero Queue Latency
- Dedicated account manager
- Eligible for Lane Stacking ? What is Lane Stacking? Add another eligible plan and your lanes add up in one account. More lanes mean more requests can run at the same time. 1× $99 = 3 Lanes 2× $99 = 6 Lanes $99 + $597 = 25 Lanes
Credits valid for 6 months
Flexible infrastructure for every stage
Start with our shared high-performance cluster for testing and side projects. When you're ready to scale, upgrade to a Dedicated Cluster Node to ensure zero-queue latency for your production AI Agents.
For testing & side projects
Get 100 Free Credits- 1 Parallel Lane †
- 100 Free Credits
- Shared Cluster Resources
For high-volume & compliance
Contact Sales- Custom Parallel Lanes †
- Dedicated Cluster Node
- SLA & Priority Support
† Parallel Lanes: Requests that can run at the same time. When one request finishes, that lane opens right away. Eligible paid plans can add lanes together for more throughput.
Latest from Our Blog
SERP API and Reader API best practices, AI development guides, and industry insights
Ready to test SearchCans?
Join thousands of developers using SearchCans for their AI applications. Check our documentation or try the API playground first.