As published on wandb.ai, June 2026. Captured by usedby on Oct 3, 2026.
What happened
Decart uses W&B Models to track every training experiment for its Oasis 3 world model, run hyperparameter sweeps, and version datasets, checkpoints, and model outputs with Artifacts. Lineage tracking and access controls give an audit trail of which data, code, and checkpoints produced the served model.
Summary written by usedby from the source page, in English. The figures are those of Weights & Biases and Decart, not ours.
Every training experiment, from early architecture explorations to final large-scale runs, was tracked in one place. Sweeps accelerated hyperparameter optimization, while Artifacts versioned datasets, checkpoints, and model outputs so any result could be traced to the exact data and configuration that produced it.
What the story claims, and what we checked
We compared the story with its live page on Oct 3, 2026.
- The passage quoted aboveCheckedCopied word for word from the page, near the name of Decart.
- The numbers in our summaryCheckedEach one is printed on the page.
- The publication dateCheckedRead from the page’s own metadata, never guessed.
- Decart uses Weights & BiasesCheckedConfirmed line. Latest check across sources: Oct 3, 2026.
- The result itselfNot checkedWe quote it; we did not measure it.




