Local Dev with CLI

You can use Agent Cloud locally on your machine to build your projects and this is the simplest way to get Agent Cloud up and running. It only takes a few minutes.

​

Setup via Docker

Learn how to get Agent Cloud running locally via Docker.

chmod +x install.sh && ./install.sh

~$ ./install.sh --help
Usage: ./install.sh [options]
Options:
--project-id ID                  Specify the GCP project ID.
--service-account-json PATH      Specify the file path of your GCP service account json.
--gcs-bucket-name NAME           Specify the GCS bucket name to use.
--gcs-bucket-location LOCATION   Specify the GCS bucket location.
-h, --help                       Display this help message.

./install.sh --project-id [INSERT PROJECT ID HERE] --service-account-json [GENERATE A SERVICE ACCOUNT IN GCP AND INSERT LOCAL PATH TO KEYFILE JSON PATH] --gcs-bucket-name [INSERT A GCP BUCKET NAME] --gcs-bucket-location [INSERT A GCP LOCATION E.G. US-CENTRAL1] --openai-api-key [INSERT KEY SK-.......]

​

System Requirements

Memory (RAM):

1. Minimum of 16GB RAM is recommended for optimal performance. However, it’s essential to allocate sufficient memory based on your specific usage.
2. Actual memory usage may vary depending on factors such as dataset size and project complexity.
3. During testing, typical memory usage ranges from 8GB to 10GB, including the operating system and other concurrently running programs within Docker.

Recommended RAM If Running Embeddings Locally:

• 🚀 Minimum of (16-32 GB) for optimal performance.
• ❌ Avoid using 8 GB.
• 💡 Consider 32 GB for excellent performance.

Disk Space:

• Minimum of 20 GB of free disk space.

Docker Resource Allocation:

• It’s crucial to allocate sufficient resources to Docker itself, especially if running AgentCloud within Docker containers.

​

Managed version

​

Troubleshooting

If you are stuck and need some help, don’t hesitate to reach out on the #help channel on discord. You can join our Discord here