Hosted platform for tracking ML experiments and managing models and datasets. Un produit fermé de Weights & Biases.
| Outil | Remplacement | Étoiles | Licence | Conditions | Auto-hébergeable | Langage | Dernière release | Dernier push |
|---|---|---|---|---|---|---|---|---|
| Trackio | Drop-in | 1.7k | MIT | Open source | Oui | Python | trackio@0.40.0 | 2026-09-30 |
| MLflow | Remplacement complet | 28k | Apache-2.0 | Open source | Oui | Python | v3.16.1 | 2026-10-06 |
| ClearML | Remplacement complet | 6.9k | Apache-2.0 | Open source | Oui | Python | v2.1.12 | 2026-09-28 |
| DVC | Partiel | 16k | Apache-2.0 | Open source | Oui | Python | 3.67.1 | 2026-10-05 |
| TensorBoard | Partiel | 7.2k | Apache-2.0 | Open source | Oui | TypeScript | 2.21.0 | 2026-08-24 |
| Aim | Partiel | 6.3k | Apache-2.0 | Open source | Oui | Python | v3.29.1 | 2026-10-06 |
| Polyaxon | Partiel | 3.7k | Apache-2.0 | Open source | Oui | MDX | v2.17.0 | 2026-09-30 |
7 alternatives
A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗
Accepts the wandb init and log calls; dashboards are stored locally or on a Hugging Face Space.
The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.
A tracking server for runs, metrics and artefacts, with a model registry.
ClearML - Auto-Magical CI/CD to streamline your AI workload. Experiment Management, Data Management, Pipeline, Orchestration, Scheduling & Serving in one MLOps/LLMOps solution
Experiment tracking, data versioning, pipelines and remote execution on one server.
🦉 Data Versioning and ML Experiments
Versions data and models alongside git and tracks experiments from the command line.
TensorFlow's Visualization Toolkit
Visualises training metrics, graphs and embeddings from local log files.
Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.
Logs and compares training runs in a local UI, without hosted collaboration.
AI Infra / AI Orchestration / AI Control Plane
Tracking, hyperparameter search and job scheduling on Kubernetes.