How many CPU cores and how much RAM does a GitHub Actions runner have?
How many cores and how much memory does ubuntu-latest have?
A standard GitHub Actions Linux runner has 2 vCPU and 8 GB of RAM on a private repository and 4 vCPU and 16 GB on a public repository, each with a 14 GB SSD; ubuntu-slim has 1 vCPU and 5 GB, and macOS on Apple M1 has 3 cores and 7 GB. More than that means a larger runner, from 4 cores and 16 GB up to 96 cores and 384 GB, for organizations on Team or Enterprise Cloud. Sengi gives 2, 4 or 8 vCPU with 8, 16 or 32 GB on any repository, on AMD Ryzen 9 7950X3D bare metal boosting past 5 GHz, from $0.003 a minute.
Last checked October 23, 2026, against GitHub’s hosted-runner reference and larger-runner specifications.
CPU and RAM by runner
| Runner | vCPU | Memory | Per minute |
|---|---|---|---|
ubuntu-latest, private repository | 2 | 8 GB | $0.006 |
ubuntu-latest, public repository | 4 | 16 GB | Not billed |
ubuntu-slim | 1 | 5 GB | $0.002 |
| GitHub 4-core larger runner | 4 | 16 GB | $0.012 |
| GitHub 8-core larger runner | 8 | 32 GB | $0.022 |
sengi-standard-2-ubuntu-2404 | 2 | 8 GB | $0.003 |
sengi-standard-4-ubuntu-2404 | 4 | 16 GB | $0.006 |
sengi-standard-8-ubuntu-2404 | 8 | 32 GB | $0.012 |
Windows and Linux arm64 standard runners match the Linux sizes; every runner, with disk and labels, is in GitHub Actions runner specs.
How do I check from inside a job?
- run: nproc && free -h && lscpu | grep 'Model name'That prints the cores, the memory and the CPU model the job landed on. GitHub does not publish its CPU model, and the jobs we measured ran on four different server CPUs. Every Sengi job reports the same AMD Ryzen 9 7950X3D, with up to 3× the multi-core and 1.65× the single-core CPU of GitHub’s 2 vCPU runner. See the runner comparison.
Frequently asked questions
Why does my job see 2 cores when GitHub says ubuntu-latest has 4?
Because the repository is private. GitHub gives public-repository jobs 4 vCPU and 16 GB, and private-repository jobs 2 vCPU and 8 GB, under the same ubuntu-latest label. Run nproc in a step to see what the job actually has.
Why did my GitHub Actions job run out of memory?
A private-repository ubuntu-latest job has 8 GB of memory shared by the runner, your build and any service containers. Parallel compiles, large test suites and Docker builds can exceed it; the step dies with exit code 137 or the job is cancelled. Lower the parallelism, or move the job to a size with more memory.
How do I get more CPU or RAM in GitHub Actions?
On GitHub, an organization on Team or Enterprise Cloud can create a larger runner, from 4 cores and 16 GB at $0.012 a minute. Sengi offers 4 vCPU with 16 GB at $0.006 and 8 vCPU with 32 GB at $0.012, on personal and organization repositories.
How much memory does a Sengi runner have?
8 GB at 2 vCPU, 16 GB at 4 vCPU and 32 GB at 8 vCPU, on AMD Ryzen 9 7950X3D bare metal. A job sees a little less than it is given, about 7 GB on the 2 vCPU size and 31 GB on the 8 vCPU size.
The fastest hardware for your GitHub Actions.
AMD Ryzen 9 7950X3D desktop CPUs boosting past 5 GHz, on bare metal with local NVMe, a fresh VM for every job, and 2 vCPU jobs ready in 2 seconds. Builds finish up to 1.9× faster than on GitHub’s runners, and at up to half GitHub’s price per minute they cost up to 74% less per build. Change one runs-on line, or give your coding agent one prompt and it opens the pull request for you.