Reference

CLI Reference

The bricqs CLI is the fastest way to manage deployments, monitor your platform, and integrate BRICQS into CI/CD pipelines.

Installation

pip install bricqs

Requires Python 3.9+. Installs the bricqs command globally.

Authentication

bricqs login

Authenticate with your BRICQS account. Credentials stored in ~/.bricqs/config.json (permissions 600).

bashbricqs login
bricqs login --email you@example.com

bricqs logout

bricqs logout

bricqs whoami

bricqs whoami

Models

bricqs models

List all deployable AI models with task type, GPU requirements, and description.

bricqs models

Deployments

bricqs deploy

Deploy a model. Polls until running by default and prints the live endpoint.

bashbricqs deploy <name> [OPTIONS]

Options:
  --model, -m TEXT    Model ID from `bricqs models`  [required]
  --env TEXT          Environment: production | preview | development  [default: production]
  --min INT           Minimum replicas (0 = scale to zero)  [default: 0]
  --max INT           Maximum replicas  [default: 1]
  --no-wait           Return immediately without polling for running state

bash# Examples
bricqs deploy my-api --model meta-llama/Llama-3-8B-Instruct
bricqs deploy staging-api --model mistralai/Mistral-7B-Instruct-v0.3 --env preview
bricqs deploy prod-api --model microsoft/phi-3-mini-4k-instruct --min 1 --max 3

bricqs status

Show all your deployments with a summary strip (active count, GPU hours, failure rate).

bashbricqs status                    # all deployments
bricqs status <deployment-id>   # single deployment detail
bricqs status --env preview      # filter by environment

bricqs logs

bashbricqs logs <deployment-id>
bricqs logs <deployment-id> --lines 200

bricqs stop

Stop a running deployment. Scales to zero — you stop being billed. The record is preserved and the deployment can be restarted from the dashboard.

bricqs stop <deployment-id>

bricqs delete

Permanently delete a deployment and its runtime. Prompts for confirmation unless --yes is passed.

bashbricqs delete <deployment-id>
bricqs delete <deployment-id> --yes   # skip confirmation (CI use)

Environment variable

Override the API base URL — useful for pointing at a staging or self-hosted BRICQS API:

bashexport BRICQS_API=https://your-api.example.com
bricqs models

CI/CD example

Deploy from a GitHub Actions workflow on every push to main:

yaml- name: Deploy to BRICQS
  env:
    BRICQS_API: https://api.bricqsai.com
  run: |
    pip install bricqs
    echo '{"access_token":"<BRICQS_TOKEN>","email":"ci@example.com"}' > ~/.bricqs/config.json
    chmod 600 ~/.bricqs/config.json
    bricqs deploy prod-api \
      --model meta-llama/Llama-3-8B-Instruct \
      --env production \
      --yes