OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Cml

CML is an open-source command-line tool for adding machine-learning development tasks to continuous integration and delivery workflows. It can run model training and evaluation, generate metric-and-visualization reports for pull requests, compare experiments across project history, monitor changing datasets and provision compute through configured runners.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Cml — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use CML when you want to run ML tasks through GitHub, GitLab or Bitbucket CI/CD and surface results in the repository’s review workflow.

Who it is for

  • Machine-learning developers who use CI/CD to run training or evaluation.
  • Teams that want experiment metrics and plots attached to code reviews.
  • Practitioners who need to allocate cloud or self-hosted compute within ML workflows.

Documented uses

  • Automate model training and evaluation when repository changes trigger CI.
  • Publish Markdown reports with metrics and plots as pull-request comments or checks.
  • Compare experiments across project history and track data or model changes through Git-based workflows.
  • Monitor changing datasets and provision machine resources for ML workflows.
Before you choose

Requirements & limits

CML is described as open source under the Apache-2.0 license and is provided as a CLI. The setup instructions require an account with GitHub, GitLab or Bitbucket; GitLab setup calls for a personal access token.

  • CML works within an existing CI/CD platform rather than providing a separate hosted ML platform.
  • The selected runner determines the compute environment; CML documents self-hosted runners and cloud allocation options.
  • The GitHub report example relies on a workflow token, while GitLab instructions specify a personal access token.
  • The optional Docker image bundles CML, DVC, Node.js and Python 3 on Ubuntu LTS.

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

Add a CML report to a GitHub Actions workflow

Run a project’s training script in GitHub Actions and post metrics or plots as a pull-request comment.

Before you start

  • A GitHub account and project repository.
  • A training script and its dependency requirements.
  • A .github/workflows/cml.yaml workflow using GitHub Actions.
  • A repository token available to the workflow.

Steps

  1. Add a workflow that checks out the repository, sets up Python and CML, and runs the project’s training command.

  2. Write metrics or plot references into a Markdown report in the workflow.

  3. Set REPO_TOKEN to the GitHub Actions GITHUB_TOKEN secret and call cml comment create report.md.

  4. Push a change and open a pull request to compare the branch; the documented example has the workflow publish the report there.

Expected outcome

A Markdown report containing project results can appear as a comment in the GitHub pull request.

Guide limitations

  • The guide describes a GitHub Actions workflow; other CI providers have separate setup instructions.
  • The project’s training code, dependencies and report contents must be supplied by the team.
Documentation 1 ↗
Evidence and transparency

Sources & review

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