# SwanHubX/SwanLab

⚡️SwanLab - an open-source, modern-design AI training tracking and visualization tool. Supports Cloud / Self-hosted use. Integrated with PyTorch / Transformers / verl / LLaMA Factory / ms-swift / Ultralytics / MMEngine / Keras etc.

Repository: https://github.com/SwanHubX/SwanLab
Canonical: https://ross.abutalabs.com/products/swanlab
Homepage: https://swanlab.cn
Language: Python
License: Apache-2.0
License Family: permissive
Topics: data-science, deep-learning, machine-learning, python, pytorch, tensorflow, transformers, mlops, model-versioning, tracking, visualization, tensorboard, logging, llm, training, ai-infra
Last push: 2026-08-26T10:30:31+00:00

## Health v2 (maintenance only)
Score: 89/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 72
- inputs: {"age_days": 1013, "days_push": 7, "days_rel": 8, "gap_med": 7, "n_releases_24m": 80}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4175, forks 216 (observed 2026-08-28T04:08:38.200438+00:00)

## What it is
SwanLab is an open-source AI training tracking and visualization platform with a Python SDK, CLI, and cloud or self-hosted dashboard. It integrates with 50+ frameworks including PyTorch, Transformers, LLaMA Factory, and Keras to log metrics, logs, and hardware usage during model training.

## Use cases
- track machine learning training experiments
- visualize training metrics like loss and accuracy
- compare experiments against a baseline
- self-host an experiment tracking dashboard
- monitor GPU and hardware usage during training
- log LLM fine-tuning runs with LLaMA Factory or verl
- find a TensorBoard alternative with team collaboration

## When to choose
- you need experiment tracking for PyTorch, Transformers, or Keras training runs
- you want a modern, self-hostable alternative to TensorBoard or Weights & Biases
- you train models in a team and need shared dashboards and experiment comparison
- you fine-tune LLMs with integrated frameworks like LLaMA Factory, ms-swift, or verl

## When to avoid
- you need general-purpose application monitoring or APM rather than ML training tracking
- your project has nothing to do with machine learning
- you only need simple plotting without experiment management

## Facets
- artifact type: library
- maturity: active
- function: logging, monitoring, data-visualization, machine-learning, llm-training, cli
- domain: machine-learning, deep-learning, data-science, large-language-models, developer-tools
- platform: python, self-hosted, cloud
- tags: mlops, experiment-tracking, tensorboard-alternative, model-training, pytorch, transformers, self-hosted, web-server, docker

## Member repositories
- SwanHubX/SwanLab (main) score 89

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:38.200438+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:22:42.027085+00:00, confidence not recorded.
  - readme: https://github.com/SwanHubX/SwanLab (fetched 2026-08-28T04:08:38.200438+00:00, sha b9bdb03a4132)
  - homepage: https://swanlab.cn (fetched 2026-08-29T09:13:21.823184+00:00, sha fd19c186ff2c)
  - registry_pypi: https://pypi.org/pypi/swanlab/json (fetched 2026-08-29T09:13:21.826687+00:00, sha 07df6919b89d)
- Data as of 2026-08-30T08:39:29.467469+00:00.