Questions, answered

No. You need a Kaggle account and an API key. Editing happens in the browser and training happens on Kaggle.

On kaggle.com, open Settings → API and create a legacy API key. Kaggle downloads a kaggle.json; paste its contents when Caudex Studio asks. It is checked with Kaggle before anything is saved.

No. It is an independent tool that talks to Kaggle's public API. It is not affiliated with or endorsed by Kaggle.

Yes, as long as the account is phone-verified — Kaggle requires that before it will give a notebook a GPU. Everything else, including the notebooks the runs create, Caudex Studio makes for you.

Runs use your own Kaggle quota. Kaggle includes free GPU and TPU hours every week, CPU runs don't touch the GPU hours at all, and the dashboard shows what is left.

The runtime you pick: CPU, an NVIDIA T4 or P100 GPU, or a TPU VM v3-8, as Kaggle provides them. The T4 runtime gives two cards, and the templates use both when the model supports it.

Kaggle ends a session at 12 hours, and keeps nothing when it does. The training templates stop themselves at 11 with the checkpoint saved, so a longer training continues in the next session instead of losing the work.

Yes. When a session stops with epochs left, Caudex Studio starts the next one for you, and keeps going until the epochs you asked for are done. If your weekly GPU hours run low it waits for Kaggle's reset rather than starting a session that would be cut off. The project page shows which session it is on, and you can stop it there.

Not yet. “Stop watching” stops following a run here, and every run links to Kaggle, where you can cancel it.

Nothing. The run is on Kaggle, not in your browser. Come back later and the log and outputs are there.

In Caudex Studio, under your Kaggle account. Each run also pushes the code to Kaggle as a new version of a notebook in your account, so it is there too.

COCO JSON, YOLO data.yaml, CVAT 1.1 XML, class folders and CursusJSON. The templates page says which template reads which.

Yes. Upload it through Caudex Studio and it becomes a Kaggle dataset in your account, which any of your projects can attach.

Nothing is deleted. You can still sign in, open everything and download it. Starting runs, creating projects and uploading datasets wait until you pay.

Caudex Studio itself: the editor, one-click runs, the live log and the outputs. Kaggle's GPU and TPU hours stay free and come from your own Kaggle account; the plan doesn't add to them.

Nothing is charged, because the trial never asks for a card, and nothing is deleted. The pricing page has what keeps working and what waits.