If your GitHub action runner bill is too high or your workflows too slows, you're in for a treat.
Introducing RunsOn:
- Prices up to 10x cheaper than official runners.
- Access to the widest and fastest range of instances available.
- On-premise, in your own AWS account.
Link👇
A common CI mistake: upgrading to a bigger runner before checking what the job is actually waiting on.
In RunsOn v3, each job gets CPU/memory/disk/network/I/O charts directly in the GitHub Actions UI by default.
No action, no OTEL backend, no CloudWatch setup.
@useblacksmith Did you maybe ran the "Before" from your own servers, with your transparent caching disabled? Because GitHub is no longer that slow I think?
@abidlabs@huggingface Nice, although self-hosted on EC2 should just use https://t.co/3A6Q8xpfbX instead 😄 All EC2 GPU instances supported, with GPU-enabled images compatible with what github expects!
Everyone seems to be hyping remote devboxes / sandboxes, but I’m still unconvinced. As developers, who is using that at scale except the big model providers? For me it’s slower hardware than my Mac, less control, more setup, and far more expensive.
In the last months I regularly discover that my tool @RunsOnRunsOn is doing 2x to 4x more builds per day across all users vs lots of VC funded and high-profile companies like Daytona, Depot, CircleCI etc. 🤯
🆕Daytona’s Agent-Native Compute: 60ms sandboxes, 50K startups in 75 sec, 850K daily runs, RL/evals, CLI > MCP, & the end of localhost https://t.co/ceZgmkIFqe
@daytonaio CEO @ivanburazin explains why AI agents need composable computers, how Daytona pivoted from human dev environments to agent sandboxes, why bare metal and stateful snapshots matter, how RL workloads went from 0% to ~50% of usage, why Kubernetes breaks down at agent scale, and why the AI cloud may look more like Stripe than AWS.
Google cloud IAM section is the worst. Can't do anything here, confusing as hell, requires permission overrides everywhere. Makes you love the AWS version.
> Actions is experiencing issues with ubuntu standard hosted runners leading to high wait times. We are actively investigating the issue
Not affecting https://t.co/3A6Q8xpfbX