Skip to main content

Hectiq Lab

Track your machine learning experiments in Hectiq Lab: runs, metrics, datasets and models, from Python or the command line.

  • Out of your way. Everything is sent in the background, and an API that is down never crashes your training.
  • Nothing to fill in. Each run records its commit, diff, packages and machine by itself.
  • Built for agents too. Every hlab command prints JSON with --json.
  • A dashboard in your terminal. hlab ui follows your runs live.

The documentation is at docs.hectiq.ai.

Installation

pip install "hectiqlab[cli]"

This gives you the Python package and the hlab command line. Other installs:

pip install hectiqlab                 # the Python package only, e.g. on a training machine
pip install "hectiqlab[tui]"          # adds the dashboard, hlab ui
pip install "hectiqlab[tui,parquet]"  # adds parquet previews in the dashboard

Quickstart

Log in once per machine, with the email and password you use on the web app:

hlab login

Then choose the project this repository records to:

hlab init hectiq-ai/demo

Now record a run:

import hectiqlab as hl

with hl.run("My first run", config={"lr": 1e-3, "epochs": 3}):
    for epoch in hl.each(range(3), "epoch"):
        loss = train_one_epoch()
        hl.add_metric("loss", step=epoch, value=loss)
    hl.models.create("checkpoints/", name="my-model")

The run ends completed, or failed with its traceback if the block raises. Find it in the web app, or with hlab run list.

Command line

hlab run list                   # the project's runs
hlab run get 42                 # one run, by its number
hlab metric get 42 loss         # a metric's values
hlab model download my-model    # a model's files
hlab launch --tag ENG-42 -- python train.py   # record a command, and say how
hlab --help                     # everything else

Add --json to any command to get JSON on stdout.

Dashboard

hlab ui

It opens on your project. Press ? to see the keys.

Settings

Variable What it does
HECTIQLAB_API_KEY Your key, instead of hlab login. Useful where the command line isn't installed.
HECTIQLAB_PROJECT The project, instead of hlab init. You can also pass project= to hl.run.
HECTIQLAB_DISABLE=1 Records nothing. In distributed training, only rank 0 records anyway.

Metadata

Release files for hectiqlab 0.5.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for hectiqlab 0.5.0
File Size Uploaded Provenance Actions
hectiqlab-0.5.0.tar.gz 348.2 kB With provenance Details for hectiqlab-0.5.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for hectiqlab 0.5.0
File Interpreter ABI Platform Actions
hectiqlab-0.5.0-py3-none-any.whl Python 3 none any Details for hectiqlab-0.5.0-py3-none-any.whl

Total release size: 600.4 kB

Release files / hectiqlab-0.5.0.tar.gz

Download URL hectiqlab-0.5.0.tar.gz
Size 348.2 kB
Tags Source
SHA-256 checksum
How to use checksums
162c7e5ba90ebca1b5450dd5b4f237b6c7dc33ad5d406bc5800ebd7fc82bf61f
BLAKE2b-256 checksum
How to use checksums
8a1335fc01da88e9d5a4ab51364cca82f10fe07a33c89e6a5cd615165f717bc2
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Release files / hectiqlab-0.5.0-py3-none-any.whl

Download URL hectiqlab-0.5.0-py3-none-any.whl
Size 252.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
2f6ad7a235b27fe9e19498645171326ddc070728e737851fc768e33a4761927f
BLAKE2b-256 checksum
How to use checksums
ad6672224a9516e90b316cbc0b4b7d03dee06f28572ac1d7009e952bd8bef780
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Security for hectiqlab 0.5.0

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.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for hectiqlab-0.5.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

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.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page