Native aarch64, 20% cheaper
arm64 runners
Native aarch64 runners at 20% below the x64 price, for teams shipping to Graviton, Ampere or Apple silicon.
No emulation
Real arm64 cores, so compilation runs at full speed.
Cheaper per minute
arm64 is priced below x64, so migrating cuts time and cost together.
Same images
The same preinstalled toolchain, built for aarch64.
If you deploy to Graviton or Ampere, building on x64 and hoping is a real risk. Native arm64 CI catches architecture-specific failures before they reach production, and usually costs less than the x64 job it replaces.
Common questions
- Is arm64 actually cheaper?
- Yes — 20% below the equivalent x64 size, and for compile-heavy workloads it is often faster as well, so the saving compounds.
- What usually breaks when moving to arm64?
- Transitive native dependencies. Node modules without an aarch64 prebuild and Python packages without a manylinux_aarch64 wheel fall back to compiling from source. Application code in Go, Rust, Java, Ruby and PHP crosses cleanly.
- Can I run arm64 and x64 in the same workflow?
- Yes, using a matrix with a different runner label per architecture. Remember to include the architecture in cache keys.
Your next build could be twice as fast, at half the price
Start free. Migrating away is the same one line, and we publish that diff too.