If your question isn't here, it probably means we haven't been asked it yet. Reach out and we'll add it.
Nope. For basic tests you just enter a URL, pick your VUs and duration, and hit run. If you do know k6 or want to write custom scripts, that's fully supported too — you get the best of both worlds.
About 90 seconds from signup to seeing results. Create an account, enter a URL, set your virtual users, and go. No configuration files, no CLI setup, no Docker containers to wrangle.
A virtual user is one simulated connection hitting your server. 100 VUs means 100 concurrent connections making requests as fast as your server can respond. It's the closest thing to real users hammering your app at the same time.
Virtual users times minutes. So a 10-minute test with 100 VUs = 1,000 VU-minutes. It's how we measure usage instead of charging per-test or per-seat, which never felt fair.
Your tests keep running — we don't cut you off mid-test. Overage gets added to your next invoice at $5 per 1,000 VU-minutes. You can also set a hard budget cap if you want a ceiling.
Yep, right from the dashboard. You keep your plan features through the end of the billing period. No cancellation fees, no hoops, no passive-aggressive 'are you sure?' modals. Ok maybe one confirmation dialog.
Not yet. We're keeping things simple for now — monthly plans, cancel anytime. If annual billing is something you'd find useful, let us know.
Real servers on AWS Fargate in us-east-1. These are actual HTTP requests hitting your infrastructure, not simulated browser traffic. Your monitoring tools will see it as real load.
Absolutely. If it responds to HTTP, you can load test it. REST APIs, GraphQL endpoints, webhooks — anything with a URL. For more complex scenarios like auth flows or multi-step sequences, use the k6 script editor.
Latency percentiles (p50, p90, p95, p99), requests per second, error rates, and error breakdowns — all in real-time while your test runs. After the test, you get per-endpoint breakdowns so you can see exactly which routes are slow.
Yes. Once a test starts running, the dashboard shows live metrics updating every second. You can also cancel a test mid-run if you've already seen what you need.
All LoadBolt traffic uses the User-Agent header "LoadBolt/1.0" and comes from a single static IP. You can allowlist either one in your WAF, CDN, or rate limiter. Check out our allowlisting guide at /guides/allowlisting for step-by-step instructions for Cloudflare, AWS WAF, Nginx, and more.
Depends on your definition of serious. If you need to simulate 3,000 concurrent users hammering your API for an hour with custom k6 scripts, absolutely. If you need geo-distributed load from 12 regions with JMeter plugins — honestly, you probably want a bigger tool. We're built for teams that need solid, straightforward load testing without the enterprise overhead.
Yes. All paid plans support team collaboration. Create an org, invite your team, and everyone can see projects, tests, and results. No per-seat charges.
You can only load test domains you've verified ownership of. This prevents abuse and makes sure you're only testing your own infrastructure. Verification is quick — just add a DNS TXT record.
The fastest way to get answers is to just try it. The free plan is genuinely free.
Start free