GPU runner

NVIDIA T4 runners for GitHub Actions

Turing-era inference GPU. Still common where cost beats clock. Cirun spins NVIDIA T4 VMs up on your cloud account per workflow — pick whichever cloud has the SKU on your existing account, CUDA + drivers come from each cloud's own Deep Learning image.

Run a workflow on NVIDIA T4 Docs

Why this fits

16 GB VRAM and ~65 TFLOPS FP16 — inference smoke tests, fine-tune sanity checks and per-PR GPU regressions.
Works with PyTorch, TensorRT, TensorFlow Serving — CUDA + drivers come from your cloud's official Deep Learning AMI / image, not your workflow.
Ephemeral by default — each job gets a clean VM, so leaked weights or hung CUDA contexts can't poison the next run.

.cirun.yml

1# NVIDIA T4 runner configuration depends on which cloud you run on.
2# See docs.cirun.io for the cloud-specific machine_image identifier.
3runners:
4  - name: gpu-t4
5    labels:
6      - cirun-gpu-t4

Drop this in your repo root. The first workflow that requests the runner label spins this configuration up on your cloud account.

Ready to run your CI here?

Cirun is free for open source. For private repos, flat monthly plans by repo count — never per CI minute.

Run a workflow on NVIDIA T4 Docs

Sources · verified 2026-05-25

NVIDIA — NVIDIA T4 datasheet