This release is a pre-release and may not be stable for production use.
Liquid Labs CLI
Command line interface for managing Liquid Labs on-prem stack.
Installation
pip install liquidai-cli
A docker-compose.yaml file is also shipped together with the package. Any changes to this file may cause some unexpected behaviors.
Run with uv
uv allows to run this tool without installing the package into the system.
uv run --directory [PATH_TO_THIS_DIRECTORY] liquidai [command] [args]
Configuration
The CLI uses a YAML configuration file (liquid.yaml) in your working directory. A default configuration will be created on first use, but you can customize it:
stack:
version: "c3d7dbacd1"
model_image: "liquidai/lfm-7b-e:0.0.1"
api_secret: "local_api_token"
# Other values will be auto-generated
database:
name: "liquid_labs"
user: "local_user"
password: "local_password"
port: 5432
schema: "labs"
Usage
Stack Management
# Launch stack
liquidai stack launch
# Launch with upgrades
liquidai stack launch --upgrade-stack --upgrade-model
# Shutdown stack
liquidai stack shutdown
# Test API endpoints
liquidai stack test
# Purge stack (removes all components)
liquidai stack purge
# Purge without confirmation
liquidai stack purge --force
Model Operations
# Run a model in docker container
liquidai model run-model-image \
--name lfm-3b-e \
--image "liquidai/lfm-3b-e:0.0.6"
# Run a HuggingFace model and expose on port 9000
liquidai model run-hf \
--name llama-7b \
--path meta-llama/Llama-2-7b-chat-hf \
--port 9000 \
--gpu-memory-utilization 0.6 \
--max-num-seqs 600 \
--max-model-len 32768
# Run a local checkpoint and expose on port 9001 to avoid conflicts
liquidai model run-checkpoint \
--path /path/to/checkpoint \
--port 9001 \
--gpu-memory-utilization 0.6 \
--max-num-seqs 600
# List running models
liquidai model list
# Stop a specific model
liquidai model stop llama-7b
# Stop a model interactively
liquidai model stop
Database Operations
# Connect to database using pgcli
liquidai db connect
Infrastructure
# Create Cloudflare tunnel with token
liquidai tunnel create --token YOUR_TOKEN
# Create tunnel interactively
liquidai tunnel create
Configuration Management
# Import configuration from .env file
liquidai config import
# Import from specific .env file
liquidai config import --env-file /path/to/.env
# Import to specific config file
liquidai config import --config-file /path/to/liquid.yaml
# Force overwrite existing config
liquidai config import --force
Command Reference
Call liquidai [command] --help to get the detailed usage reference.
Metadata
Release files for liquidai-cli 0.0.1b0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| liquidai_cli-0.0.1b0.tar.gz | 14.2 kB | Without provenance | Details for liquidai_cli-0.0.1b0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| liquidai_cli-0.0.1b0-py3-none-any.whl | Python 3 | none | any | Details for liquidai_cli-0.0.1b0-py3-none-any.whl |
Total release size: 31.2 kB
Release files / liquidai_cli-0.0.1b0.tar.gz
| Download URL | liquidai_cli-0.0.1b0.tar.gz |
|---|---|
| Size | 14.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7d6c10cfba0b59e38f7d223c1061cda52ed4ee36b6184dc5e5e3cf314dbbfddc
|
|
BLAKE2b-256 checksum How to use checksums |
1dc2ef3e853663d084262c3dd191fcdac09daa1321bb9bbd63ddb47684792625
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.17
|
Release files / liquidai_cli-0.0.1b0-py3-none-any.whl
| Download URL | liquidai_cli-0.0.1b0-py3-none-any.whl |
|---|---|
| Size | 16.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
32844ef4cf6cf8bf337101f2815762934da4dde782c794d0118332d71c92974f
|
|
BLAKE2b-256 checksum How to use checksums |
aa23963099b084528bd1010bbf3b7f85c426cf2b90b2317f7d491917a0f46480
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.17
|
Security for liquidai-cli 0.0.1b0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.