Compute-heavy workloads. Dedicated cores.
CPU-Optimized plans give you 100% guaranteed dedicated vCPU threads — never shared, never throttled. Built for CI/CD, batch processing, video encoding, and game servers — starting at $24/mo.
Cores you can count on,
wall-clock to wall-clock.
CPU-Optimized strips out every variable that hurts compute throughput. What you measure in dev is what you get in production.
100% dedicated vCPU
Every vCPU thread is yours alone — never oversubscribed. No noisy neighbours, no contention, no surprises under load.
Sustained turbo clocks
Boost frequencies hold under sustained load. Your batch jobs and video encodes finish in predictable time, every run.
Zero CPU steal
Bare-metal-class isolation. No hypervisor steal time stealing your throughput — verified hourly across the fleet.
NVMe SSD storage
Pure NVMe block storage with up to 7,000 MB/s sequential reads. Container layers and build caches load instantly.
20 TB bandwidth
Generous outbound allowance on every plan. Inbound traffic is always free. Overage billed at $1/TB.
DDoS protection — free
Enterprise-grade DDoS mitigation included on all plans at no extra charge. Filters up to 1 Tbps of volumetric attacks.
Pay for cores, not for surprises
Every plan includes 100% dedicated vCPU threads, NVMe SSD storage, and 24/7 support. No oversubscription, ever.
Questions? Compare all plans →
Built for the workloads that count cycles.
Builds finish in predictable wall-clock time.
Self-hosted GitHub Actions, GitLab Runners, Buildkite agents, and Jenkins workers run faster and cost less than cloud-vendor minutes — with zero CPU steal.
- Run unlimited parallel builds at fixed cost
- No oversubscription = no flaky timing tests
- 20 TB bandwidth covers cache pulls
Push pixels, not the meter.
FFmpeg, x264/x265, and AV1 encodes run at full turbo. Predictable per-job time means deterministic batch costs for transcode pipelines and thumbnail generation.
- Sustained 3.7 GHz turbo on all cores
- AVX2 / AVX-512 acceleration available
- Pair with object storage for output
Tight tail latency, every tick.
Counter-Strike, Minecraft, Rust, and custom UDP game loops thrive on dedicated cores. Predictable frame-time variance keeps players from rage-quitting.
- Sub-millisecond same-DC latency
- 1 Gbps uplink with anti-DDoS shielding
- Snapshot-based world rollback in seconds
Frequently asked questions
Dedicated cores, zero contention.
100% dedicated vCPU, zero contention. Cancel anytime.