A year ago, Lagless was a single node in a Dallas datacenter. One rack, one location, two founders, and a conviction that game hosting didn't have to be the mess we'd watched it become.
I'm writing this from a different vantage point. We're 100+ nodes deep, in 8 global locations, with over 3,000 active servers, and a team of 14 people who actually care about the product they're building. That's not a brag — it's context for the rest of this post. I want to walk you through what 2025 actually looked like, where Q1 2026 has taken us, and where things get interesting from here.
What 2025 actually was
We launched in February 2025 from Dallas — the flagship location for Elcro Digital Services, our parent company. One node. The plan was honestly modest: build the host we always wanted to use, prove the model worked, expand carefully.
That's not what happened.
By the end of 2025 we were running 100+ nodes across Dallas, Atlanta, Los Angeles, Miami, New York, Singapore, London, and Frankfurt. We crossed 1,000 active clients before our first birthday. The scale curve was steeper than anything we'd modeled, and a lot of late nights were spent making sure quality didn't slip while volume climbed. Some of them I lost. Most of them, I'm proud of.
A few things we shipped along the way that mattered:
We held to flagship-tier hardware as the standard, not the upsell. AMD Ryzen 9950X became our primary chip across most locations, with equivalent silicon (9900X, 4545P, Intel Ultra 7 265) filling specific roles where it made sense. No mystery hardware. No "premium tier" gating real performance behind a paywall.
We built our DDoS posture around picking the right provider for each location rather than blanket-applying one vendor globally. GSL and Cosmic protect different parts of our footprint — the choice was deliberate, region by region. That's the kind of decision that doesn't show up in marketing copy but absolutely shows up the night someone tries to throw a UDP flood at your server.
And the 60-second deploy promise — the one we put on the homepage and the one we get asked about most — held. Not as a slogan. As a measurable fact across the year.
We also added Hytale to the platform 3 months ago. That one was a bet on a community we believed in, and it's already paying back in interest.
Q1 2026 — and a little bit of Q2
We crossed 3,000 active clients this quarter. To put that in scale: we tripled our customer base in roughly five months. That's the part nobody on the team actually predicted. We were prepared for growth — not for that growth.
Real talk: scaling that fast forces you to either raise prices, cut corners, or get smarter. We picked the third one, and a lot of this update is about how.
We grew the team from 3 people at founding to 14 today. Engineering, support, infrastructure, the people who actually pick up the chat when you have a question at 2am. Hiring slowly is fashionable advice. We hired deliberately. There's a difference.
The thing nobody in this industry wants to talk about
If you're running a server right now, you've probably noticed RAM prices doing something weird. The DDR5 supply situation is genuinely bad — production constraints, datacenter demand, the AI buildout sucking memory out of the consumer and prosumer market. Hosts everywhere are absorbing it, passing it along, or quietly downgrading what they advertise.
We're not raising your price.
That's not a marketing line. It's a strategic decision we're spending real money on. The way we're keeping it that way is by leaning hard into a hybrid hardware model — owned where it makes sense, rented from reliable regional vendors where it makes more sense, every node held to the same spec floor regardless of which side of that line it sits on. We were already running this model in some locations. We're now running it in most.
It's a less glamorous answer than "we own everything." But owning everything in this market means either eating margin until you can't, or breaking the promise you made on the homepage. Renting smartly from vendors we've vetted, with hardware we've spec'd ourselves, lets us keep both pricing and quality stable through a market that isn't.
If we ever have to change pricing, you'll hear it from me directly. Not buried in a billing email.
What's next
We're dropping a new location.
I'm not going to give you a date because I'd rather under-promise and over-deliver, but Amsterdam is coming online soon — our 9th location and our second European point of presence. EU customers should expect tighter pings, redundancy if Frankfurt has a bad day, and the same hardware standards we hold everywhere else.
Beyond Amsterdam, we're investing in three things over the next two quarters:
Infrastructure. More capacity, more redundancy, more headroom. The growth curve we just hit isn't slowing down, and we'd rather be 6 months early on capacity than 6 days late.
Team. We'll keep growing carefully. Support hours are expanding, engineering is expanding, and we're bringing on people who actually play the games they support — because "real humans, not scripts" was a promise we made and one we intend to scale with.
You. Genuinely. The most useful product feedback we got in 2025 came from customer conversations — Discord threads, support tickets, replies to emails I sent expecting silence. Keep telling us what's broken. Keep telling us what's missing. We read it. We act on it. That's not a value statement; it's how we've shipped most of what we're proud of.
The short version
100+ nodes. 8 locations soon to be 9. 3,000+ servers running. 14 people. A market that's testing every host's nerve, and a strategy we're confident enough in to put in writing.
A year ago we said we were building the host we always wanted. We meant it then. We mean it now.
— Lev Founder & CEO, Elcro Digital Services LLC dba Lagless.gg