Synthient tracked providers
Includes reseller brands. Some tags require restricted access.
Synthient vs IPQualityScore
Synthient vs IPQualityScore. Name the proxy service.
Synthient tracks 2,500+ providers, including reseller brands. Each IP response combines named services, last-seen timestamps, observed behavior, device data, and a 0–100 risk score.
Why choose Synthient
Coverage and delivery.
Synthient focuses on IP infrastructure and provider attribution. IPQualityScore combines IP reputation with email, phone, and other fraud checks.
Synthient Parquet snapshots
Hourly exports and daily rollups. Feed access required.
Synthient IPs / batch request
Enrich up to 1,000 IPv4 or IPv6 addresses in one request.
Synthient vs IPQualityScore
API and data comparison.
Response fields, delivery formats, and access requirements.
| Capability | Synthient Infrastructure intelligence | IPQualityScoreDocumented offering |
|---|---|---|
| IP lookup | GET /api/v4/lookup/ip/{ip_address} with x-api-key authentication. No browser user-agent input. | API key and IP in the URL path. Optional user_agent, user_language, and strictness inputs. |
| Risk signals | intelligence.risk_score is 0–100, with providers[], behavior, and categories returned separately. | fraud_score is 0–100, with proxy, vpn, tor, bot_status, and recent_abuse. Request context and strictness can affect scoring. |
| Mobile detection | network.type identifies the network. providers[].type can identify MOBILE_PROXY. devices[].os describes an IP-associated observation, not this visitor’s browser. | mobile comes from user_agent and is false when that input is absent. connection_type separately classifies networks as Residential, Mobile, or Data Center, among other types. |
| Provider and reseller attribution | providers[] returns named services and resellers, proxy type, and last_seen in Unix seconds. | proxy, vpn, and tor classify traffic. active_vpn and active_tor indicate activity, not a last-seen timestamp. |
| Data access | Batch lookup supports 1,000 IPs. Parquet feeds and NDJSON Firehose are separate access options. | The Proxy & VPN Detection API returns an IP risk response with a success flag and error details. |
| Scoring controls | Risk score, named providers, observation times, and behavior are separate fields available to your rules. | strictness and allow_public_access_points adjust detection sensitivity. Higher strictness can increase false positives. |
Provider counts are not accuracy benchmarks. Features and freshness depend on the plan.
Using the data
Investigate what produced the score.
A high score alone does not name a service. Synthient returns provider identities and observed activity alongside the score.
Name the proxy service
providers[] gives analysts a service name, proxy type, and last-seen time. Reseller tags distinguish the brand selling access from the network hosting the exit.
See observed activity
behavior can include CREDENTIAL_STUFFING and ACTIVE_CRAWLER observations from a 90-day window. These describe activity associated with the IP, not proof about the current visitor.
Review exits in bulk
Enrich up to 1,000 IPs per batch request, or query Parquet proxy observations in your warehouse. Group results by provider tag to review the addresses observed under the same service.
Before you decideQuestions,
Questions,
answered.
Why retain reseller tags instead of just a proxy flag?
A proxy flag does not name the service. providers[].provider identifies the observed provider or reseller; type and last_seen describe the observation. Analysts can group rotating exits by that service tag.
Can I use the same risk threshold in both products?
No. Both use 0–100, but the scores are not calibrated alike. Test thresholds on labeled traffic; neither number is a probability that this visitor is committing fraud.
Does Synthient replace IPQS email and phone validation?
No. Synthient specializes in IP and infrastructure intelligence. Keep the email, phone, and transaction checks your workflow needs.
How should I migrate from IPQS?
Use the migration guide for the endpoint and authentication change. Map fields rather than just renaming them: active_vpn is not last_seen, and mobile is not an observed operating system. Test risk thresholds separately.
Build rules around the provider, not just the score.
Bring a labeled traffic sample. Compare Synthient’s reseller tags and observed activity with your current IPQS results.