Coverage
Network coverage
Every market runs its own ASN footprint, edge latency, and compliance profile. Open one to see what we actually cover.
| Code | Country | Continent | Tier-1 ASNs | Top metros | |
|---|---|---|---|---|---|
| au | Australia | Oceania | 8 | Sydney · Melbourne · Brisbane | → |
| ca | Canada | North America | 8 | Toronto · Montreal · Vancouver | → |
| fr | France | Europe | 8 | Paris · Lyon · Marseille | → |
| de | Germany | Europe | 8 | Berlin · Munich · Frankfurt | → |
| jp | Japan | Asia | 8 | Tokyo · Osaka · Kyoto | → |
| nl | Netherlands | Europe | 8 | Amsterdam · Rotterdam · The Hague | → |
| sg | Singapore | Asia | 8 | Singapore · Jurong · Tampines | → |
| kr | South Korea | Asia | 8 | Seoul · Busan · Incheon | → |
| gb | United Kingdom | Europe | 8 | London · Cambridge · Edinburgh | → |
| us | United States | North America | 8 | New York · San Francisco · Seattle | → |
Exit classes
Every class of exit, on one API
The four exit classes that matter — and one (datacenter) that still matters for the right jobs.
4G Mobile Proxies
Real carrier LTE sessions with live SIMs and rotating CG-NAT public IPs. Used by AI teams testing mobile-first response paths, cellular content policy on model APIs, and regional carrier-level variation.
- ›Pool type: Live carrier SIMs
- ›Countries: 10 (focused)
- ›Carriers: Tier-1 per country
5G Mobile Proxies
5G SA sessions with IPv6 dual-stack and sub-30ms RAN latency. Narrow but real — used by AI teams that need eval paths matching the mobile networks their downstream product will deploy on.
- ›Pool type: Live 5G SA + NSA SIMs
- ›Countries: US, UK, DE, JP, SG (expanding)
- ›IP version: IPv4 + IPv6 dual-stack
Datacenter Proxies
Unmetered bandwidth, deterministic IPs, sub-10ms latency to the nearest edge. The SquadProxy core product for bulk training-corpus collection, open-source dataset downloads, and evaluation against non-hostile AI research infrastructure.
- ›IP allocation: Shared pool or dedicated
- ›Latency: < 10ms to regional AI edge
- ›Bandwidth: Unmetered (dedicated)
ISP Proxies
Datacenter-hosted IPs announced under real residential ASNs. Stable, persistent, and the right tool when cookies and session state need to survive across a multi-turn evaluation or a sustained RAG ingestion.
- ›IP allocation: Static, dedicated
- ›ASN class: Residential-announced
- ›Latency: < 50ms regional
Residential Proxies
Real home broadband across our 10 focus countries with district-level routing where it matters. Used by AI teams for regional model evaluation, RAG ingestion from geoblocked sources, and corpus diversity sampling.
- ›Pool size: 4M+ IPs
- ›Countries: 10 (focused)
- ›Protocols: HTTP, HTTPS, SOCKS5
What SquadProxy is
A proxy network built against AI workloads, not a general-purpose scraping product pivoted to AI
The commercial proxy category has existed for a decade, shaped by the needs of e-commerce scraping, SEO tracking, and ad-verification workflows. AI data collection is a different shape of problem — training-corpus pulls at TB scale, regional LLM evaluation across 10+ countries, RAG ingestion from sources that geoblock cloud ASNs, safety red-team workloads that cross jurisdictional lines by design. General-purpose proxy products were built for the older shape; they serve the newer one as a vertical.
SquadProxy is the reverse. Every product decision — the five exit classes, the per-request header-based class routing, the metered-residential + unlimited-datacenter plan shape, the 10-country focus with deep ASN instrumentation rather than 195-country surface-level coverage — is optimised for the AI workload surface first.
One gateway
Header-based class routing, not per-class endpoints
Your ingestion pipeline switches between residential, ISP, datacenter, and mobile exits via a header on one endpoint. Per-source routing stays in your code, not in your proxy configuration. See the routing matrix for the patterns AI pipelines end up on.
Provenance
Residential pool built on opt-in SDK integration
Peers receive value (SDK features, rewards, ad-free tiers) in exchange for bandwidth when their device is idle. No stealth installations, no bundleware, no children's devices. For research that publishes its data sources, this matters. More in the provenance post.
Economics
Residential metered, datacenter unlimited
Residential costs scale with volume; datacenter doesn't. Every plan prices both honestly. You pay for the traffic through real home broadband and get cloud bandwidth as part of the tier. See the pricing page for the full plan shape.
Focus
10 countries instrumented deep, not 195 surface-level
The United States, United Kingdom, Germany, France, Japan, Netherlands, Canada, Singapore, South Korea, and Australia. ASN depth that matters for regional AI work, with latency-measured edges. See the network coverage.
Recent writing
From the SquadProxy research engineering team
Long-form notes on proxy selection, AI data collection, RAG infrastructure, evaluation methodology, and the parts of the internet we spend our days routing traffic across. The full blog has everything; four recent ones below.
23 Apr 2026 · Nathan Brecher
Proxies for AI browser agents: the 2026 workload shape
AI browser agents — Claude Computer Use, ChatGPT Operator, Gemini-based web agents, and the long tail of open-source browser agents — hit the web from a new angle. The proxy layer below them has specific requirements that general- purpose proxy products don't meet. Here's what matters.
23 Apr 2026 · Nathan Brecher
Proxies for ChatGPT Operator: browser-agent configuration that works
ChatGPT Operator runs browser-based task execution for end users. Operator-style agents (Operator itself, open-source clones, custom GPTs with browsing) all share a proxy configuration shape. A working reference.
23 Apr 2026 · Nathan Brecher
Proxies for Claude Computer Use: session patterns and exit-class choice
Claude Computer Use operates browsers at screen-pixel level. The proxy layer below it needs specific configuration to keep agent sessions coherent and avoid anti-bot challenges. A working guide for production Computer Use deployments.
22 Apr 2026 · Nathan Brecher
Choosing a proxy for LLM training data collection: criteria that actually matter
Listicles ranking "best proxies for AI" miss the criteria that AI engineers weigh in practice. A honest breakdown of the tradeoffs — pool provenance, ASN diversity, bandwidth economics, concurrency ceilings, and legal footprint — for teams collecting LLM training data at scale.
Workloads we build for
Five AI workload shapes
Each of these has its own routing pattern, session shape, and concurrency envelope. The links below are use-case pages with the full technical framing; the blog has the deeper methodology writeups.
Pricing
Transparent, metered pricing
Pay for bandwidth, not per-port — the way every honest proxy business should be priced. Every plan gives access to all 5 exit classes and all 10 focus countries; the allocations below are the included monthly quotas. Annual billing saves 20%.
Solo
For individual researchers running evaluation scripts and prototype RAG pipelines.
$149/ month
or $1,430/year (save 20%)
50 GB residential · unlimited datacenter · 200 concurrent sessions
- ✓Access to all 5 exit classes · 10 focus countries
- ✓50 GB residential · unlimited datacenter
- ✓5 static ISP IPs · 5 GB 4G mobile
- ✓1 seat · 200 concurrent sessions
- ✓Python + Node SDK + REST API
- ✓Per-request metering (not time-based)
- ✓Email support (24h response, business days)
- ✓Overage: $3/GB residential · $6/GB mobile
Best for
- Solo researchers
- Evaluation scripts
- Prototype RAG
Team
Most popularFor AI startups and mid-size labs splitting capacity between training and evaluation.
$699/ month
or $6,710/year (save 20%)
500 GB residential · unlimited datacenter · 1,000 concurrent sessions
- ✓Access to all 5 exit classes · 10 focus countries
- ✓500 GB residential · unlimited datacenter
- ✓25 static ISP IPs · 25 GB 4G mobile
- ✓10 seats ($29/mo per extra seat) · 1,000 concurrent sessions
- ✓City-level geo-routing + ASN targeting
- ✓99.9% uptime SLA
- ✓Priority Slack support (4h response, business hours)
- ✓Python + Node SDK + REST API + webhooks
- ✓Overage: $3/GB residential · $6/GB mobile
Best for
- AI startups
- Mid-size labs
- Model eval teams
Lab
For academic labs, eval consortia, and frontier model companies running sustained workloads.
$2,999/ month
or $28,790/year (save 20%)
2 TB residential · unlimited DC · 50 GB 4G + 20 GB 5G · 3,000 concurrent sessions
- ✓Access to all 5 exit classes · 10 countries on 4 continents
- ✓2 TB residential · unlimited datacenter
- ✓100 static ISP IPs · 50 GB 4G + 20 GB 5G mobile
- ✓50 seats ($19/mo per extra seat) · 3,000 concurrent sessions
- ✓Dedicated gateway lane (bypasses shared-pool queues on us-east-1 + eu-west-1)
- ✓99.95% uptime SLA
- ✓Dedicated Slack channel (1h response, business hours)
- ✓Custom BGP prefix on request (additional fees apply)
- ✓Overage: $2.50/GB residential · $5/GB mobile
Best for
- Academic labs
- Large eval consortia
- Frontier model companies
Enterprise
Custom contracts with dedicated infrastructure, volume pricing, and research-grade SLAs.
Custom pricing
Custom (from 5 TB/mo residential) · unlimited concurrent sessions
- ✓Volume pricing from 5 TB/mo residential
- ✓Dedicated BGP prefix + ASN announcement
- ✓Unlimited concurrent sessions · unlimited seats
- ✓99.99% uptime SLA with financial credits
- ✓Named Technical Account Manager + 24/7 on-call paging
- ✓Custom AUP, DPA, on-site deployment option
- ✓Research / academic discount (30–50% off Team or Lab)
- ✓Annual contract · wire, ACH, USDC/USDT/BTC settlement
Best for
- Frontier labs
- Eval consortia
- Enterprise AI
All plans include 14-day refund, single endpoint with regional failover, HTTP(S) + SOCKS5 on every exit class, access to all 5 exit classes and all 10 focus countries, and Python + Node SDKs. Concurrent sessions = simultaneous TCP sessions through the gateway. Overage warnings fire at 80% and 100%; traffic continues only if overage billing is enabled on your account.
FAQ
Frequently asked questions
Are your proxies legal to use?
The proxies themselves are legal in every jurisdiction we operate. What you do through them is your responsibility — circumventing technical access controls, violating T&Cs, or automating regulated services (banking, gambling, healthcare) may expose you to civil or criminal liability. Our AUP lists hard-no targets; everything else is yours to evaluate.How are you different from Bright Data, Oxylabs, Smartproxy?
SquadProxy runs a smaller, ASN-curated pool with honest sourcing. You won't see 100M+ IP claims because we've audited what we sell. In exchange you get tighter ASN mixes, fewer cooked subnets, and engineers who actually know the network.What's the difference between residential and ISP?
Residential proxies are sessions on real home broadband, rotating per request or per sticky window — priced per GB. ISP proxies are static IPs announced under residential ASNs from our own infrastructure — priced per IP per month. Use residential for breadth and diversity, ISP for persistence and latency.Do you support IPv6?
Yes — on our 5G mobile and select residential pools (notably JP, US T-Mobile). IPv4 remains the default everywhere; request IPv6 via the X-Squad-Stack: v6 header on supported pools.How does billing work?
Rotating residential and mobile are billed on outgoing bandwidth (GB). ISP is billed per dedicated IP per month, bandwidth unmetered. Datacenter is either shared-bandwidth or dedicated-IP/unmetered depending on plan. Overages are charged at the plan's per-GB rate; you'll see warnings at 80% and 100%.What's your refund policy?
We offer a 14-day refund on first-month subscriptions if less than 10% of bandwidth has been consumed. See our refund policy page for the full terms.