FAQ
Frequently asked questions about NeoRun.
General
What is NeoRun?
NeoRun is an AI-powered platform that takes a GitHub repository, analyzes it with AI, builds it in a Docker container, and gives you a running URL — all in minutes. No Dockerfile or deployment config required.
Which languages are supported?
NeoRun supports any language that can run in a Docker container, including:
- Python (Flask, FastAPI, Django, Streamlit, Gradio)
- JavaScript/TypeScript (Node.js, Next.js, React, Vue, Express)
- Go, Rust, Java, Ruby, and more
If your project has a Dockerfile, NeoRun will use it directly.
Do I need a Dockerfile?
No. NeoRun’s AI agent analyzes your repository and generates the build configuration automatically. However, providing a Dockerfile gives you full control over the build.
Is NeoRun free?
Yes, there is a free tier with:
- 3 concurrent pods
- 1 GPU pod (4GB VRAM)
- 2GB RAM per pod
- 15-minute max build time
- 30-minute idle timeout
How does the AI analysis work?
NeoRun uses Google Gemini to:
- Identify your project’s language and framework
- Detect dependencies and entry points
- Determine the correct port and start command
- Generate a Docker build plan
- Run safety checks on the repository
Deployments
How long does a deployment take?
Most deployments complete in 2–5 minutes. Complex projects with large dependencies may take up to 15 minutes.
Can I deploy private repositories?
Yes. Connect your GitHub account and NeoRun can access your private repos. You can also deploy via direct URL with a GitHub token.
Can I redeploy after code changes?
Yes. Click the Redeploy button on your project page, or use the CLI:
New-NeoRunDeployment -ProjectId "your-project-id"What happens when a build fails?
NeoRun shows detailed build logs with the failure point highlighted. You can view the AI analysis, dependency resolution, and Docker build output to diagnose the issue. See Troubleshooting for common fixes.
Pods
What is a pod?
A pod is a running instance of your application inside a Docker container. Each pod gets a unique URL where your app is accessible.
How long do pods run?
Pods run until you stop them or they hit the idle timeout. Free tier pods are stopped after 30 minutes of no traffic. Paid plans have configurable timeouts.
Can I SSH into my pod?
NeoRun provides a web terminal for accessing your running pod. Navigate to your pod and click the Terminal tab.
What resources does each pod get?
| Resource | Free Tier | Pro |
|---|---|---|
| CPU | 1 core | 4 cores |
| RAM | 2 GB | 8 GB |
| Storage | 5 GB | 20 GB |
| GPU VRAM | 4 GB | 16 GB |
| Network | Shared | Dedicated |
GPU
Which GPUs are available?
NeoRun uses NVIDIA GPUs. The free tier provides access to GPUs with up to 4GB VRAM. Paid plans support T4 (16GB), A10G (24GB), and A100 (40GB/80GB).
Do I need to install CUDA drivers?
No. NeoRun’s GPU containers come with CUDA pre-installed. Use a CUDA base image in your Dockerfile, and NeoRun handles the runtime.
Can I run LLMs on NeoRun?
Yes! Check the GPU Guide and the template gallery for ready-to-deploy LLM configurations including Ollama, text-generation-webui, and more.
Security
Is my code secure?
Yes. Each build and pod runs in an isolated Docker container with:
- Network isolation between containers
- Seccomp system call filtering
- Read-only root filesystem (where possible)
- No host filesystem access
- Resource limits enforced
Are environment variables secure?
Secrets are stored encrypted using Supabase Vault and are only injected into your container at runtime. They are never logged or visible in build output.
Billing
How does billing work?
Free tier is free forever with the limits above. Paid plans are billed monthly based on resource usage (compute hours, GPU hours, storage).
What happens if I exceed my quota?
New deployments will be blocked until you’re within quota. Existing pods continue running. You’ll see a notification in the dashboard.