dvclive
Experiments logger for ML projects.
Decision gist · record as of 2026-08-14
Yes, if you are already using DVC or planning to adopt it for experiment management. DVCLive is a natural fit for teams that want lightweight, Git-friendly metric logging without external services. If you are not using DVC and prefer a standalone logger, the nine runtime dependencies and DVC integration requirement make it less attractive. No known security vulnerabilities and active maintenance support the recommendation.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.9 or later.
- A DVC repository (git init && dvc init) is expected for full functionality.
- Low friction install with a pure-Python wheel.
License · maintenance · safety
Apache-2.0 (permissive) — Apache-2.0 permissive license allows commercial and private use with minimal restrictions.
last release 2026-06-05 (70 days) · last repo commit 2026-08-10 · 196 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 569,757 downloads/mo, #5,959 on PyPI
Alternatives
Verify before relying
pip install dvclive
from dvclive import Live
with Live() as live:
live.log_param("learning_rate", 0.002)
live.log_metric("accuracy", 0.5)
live.next_step()- Whether nvidia-ml-py is required for all use cases or only when GPU metrics are needed
- Performance characteristics when logging high-frequency metrics in long-running experiments
What it is and what it does
DVCLive is a Python library that logs machine learning experiment metadata—metrics, parameters, and plots—to plain text files compatible with DVC. Unlike heavier ML logging platforms, it stores everything locally as versioned files that can be tracked by Git or DVC, eliminating the need for external services or servers. It integrates with the DVC ecosystem to enable experiment comparison and visualization through the DVC CLI, VS Code extension, or DVC Studio.
The library is designed for teams already using DVC for data versioning and experiment management. It provides a simple context-manager API for logging during training loops, supports framework integrations, and can save experiment runs as hidden Git commits for reproducibility. The main trade-off is that it assumes you're working within a DVC-managed repository rather than offering a standalone logging solution.
Use it for
- Log training metrics and parameters during model development, then compare across multiple experiment runs using dvc exp show.
- Track validation metrics in a Git-versioned format without setting up external services.
- Integrate with PyTorch Lightning or Scikit-learn workflows to automatically capture metrics alongside your training code.
- Visualize experiment results in VS Code using the DVC extension or push to DVC Studio for team collaboration.
- Store plots and metrics as plain text files that can be reviewed in version control history without external dependencies.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are already using DVC or planning to adopt it for experiment management.
DVCLive is a natural fit for teams that want lightweight, Git-friendly metric logging without external services. If you are not using DVC and prefer a standalone logger, the nine runtime dependencies and DVC integration requirement make it less attractive. No known security vulnerabilities and active maintenance support the recommendation.
Install
dvclive on PyPI
Before you install
Low friction install with a pure-Python wheel. Active maintenance (last commit 2026-08-10) and recent release (2026-06-05). Depends on nine runtime packages including dvc, dvc-render, and dvc-studio-client, which may add setup complexity if not already present.
Requires Python 3.9 or later. A DVC repository (git init && dvc init) is expected for full functionality.
License in practice
Apache-2.0 permissive license allows commercial and private use with minimal restrictions.
Quickstart
pip install dvclive
from dvclive import Live
with Live() as live:
live.log_param("learning_rate", 0.002)
live.log_metric("accuracy", 0.5)
live.next_step()Verify before relying
- Whether nvidia-ml-py is required for all use cases or only when GPU metrics are needed
- Performance characteristics when logging high-frequency metrics in long-running experiments
Package facts
| License | Apache-2.0 permissive |
| Python support | Supports the current Python release >=3.9 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 9 packagesdvcdvc-renderdvc-studio-clientfuncygtoruamel.yamlscmrepopsutilnvidia-ml-py |
| Maintenance | Actively maintained 70 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 569,757 / month, #5,959 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Programming Language :: Python :: 3.14Programming Language :: Python :: 3.9 |
Evidence: dvclive-3.49.1-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “machine learning logger”
- dvcliveDVCLive logs machine learning metrics, parameters, and plots to…
- neptune-scaleNeptune-scale is a Python client library for logging and monitoring…
- spacy-loggersProvides logging integrations for spaCy model training, supporting…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also dvc · dvc-studio-client · dvc-render · azureml-mlflow · neptune · dvc-azure · trackio · traceml · neptune-scale · spacy-loggers