OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Evaluation & Observability

Labml

Labml is an open-source Python developer tool for recording deep-learning experiment metrics and context, viewing running experiments from a phone or laptop, and monitoring computer hardware usage.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Labml — OpenViST project profile in AI Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Labml when you want to monitor model-training progress and hardware usage, particularly when a project can integrate its training loop with the Python tracker and connect to an experiments server.

Who it is for

  • Deep-learning developers who want to inspect training runs while they execute.
  • Teams that need experiment metrics stored alongside configurations, hyperparameters and Git commit information.
  • Developers who want to check training-computer hardware usage.

Documented uses

  • Record values such as loss and accuracy from a training loop.
  • Keep experiment configurations, hyperparameters and Git commit details with run records.
  • View active experiments from a mobile phone or laptop.
  • Start hardware monitoring on a computer with the Labml command.
Before you choose

Requirements & limits

  • Experiment tracking uses a Python package, a project-level .labml.yaml configuration and an experiments server.
  • The documented server setup requires MongoDB and can be accessed through localhost or a server IP.
  • Hardware monitoring is a separate command and lists psutil and py3nvml among its dependencies.
  • The repository identifies Python as its primary language and MIT as its license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up experiment tracking with a Labml server

Connect a Python project to a Labml experiments server and record training metrics.

Before you start

  • MongoDB installed for the server
  • Python environment with pip
  • A training project and metrics to record

Steps

  1. Install the server package with the command below and start it with labml app-server.

    Documented command
    pip install labml-app

  2. Install the project package with the command below.

    Documented command
    pip install labml

  3. Create .labml.yaml in the project root and set app_url to the server's /api/v1/default address.

  4. Use experiment.record and tracker.save in the training loop to record metrics.

Expected outcome

The project is configured to send experiment metrics to the Labml server, where runs can be viewed through its interface.

Guide limitations

  • The README's integration example uses PyTorch.
  • The server setup requires MongoDB; the guide does not document a managed-hosting option.
Documentation 1 ↗
Source-based practical guide

Start hardware-usage monitoring

Install the listed dependencies and start Labml's hardware monitor on a computer.

Before you start

  • Python environment with pip

Steps

  1. Install the listed packages with the command below.

    Documented command
    pip install labml psutil py3nvml

  2. Start monitoring with labml monitor.

Expected outcome

Labml starts its hardware-usage monitoring command on the computer.

Guide limitations

  • The README does not specify which hardware metrics are displayed.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
labmlai/labml · Official homepage title · retained in its source language
How we review sourcesSuggest a correction