- Go to AI Universe → AI Chatbot/Search → Training Center.
- At the top you see the numbers: Total, Pending, Processing, Completed, Failed and Tokens (how much AI was used).
- Below is the list of items. Use the filters to show only one source or one status.

- Pending – waiting to be learned.
- Processing – being learned right now.
- Completed – learned. AI Search can use it.
- Failed – something went wrong.
- Select the item (for example a Failed one).
- Click Re-queue. The item goes back to Pending.
- Click Run Task Now to start training.
- Select the item.
- Click Delete. AI Search will no longer use this content.
Scheduler
Training runs in the background with the TYPO3 Scheduler (the TYPO3 tool that runs tasks automatically at set times).- Each website has one task called T3CS Training (Site N). N is the page ID of your website’s start page.
- AI Search creates this task for you when you save your first data source.
- To start training right away, click Run All on the Dashboard or Run Task Now in the Training Center.
- Reads the sources again that you marked with Sync.
- Trains all new content (status Pending).
- Deletes old Failed items after the number of days in Retention Days. Trained content is kept.
Sync only marks content to be read again. The AI learns it when the training task runs.
- Go to the Scheduler module.
- Create a new task and choose Execute console commands.
- Choose
t3af:history:cleanup. - Enter the number of days to keep (default: 90).
- Set how often it runs on the Timing tab.
- Save.

Training settings
How content is split, embedded and cached is set on the shared Training card (AI Universe → AI Foundation → AI Features → Training, badge AI Chatbot/Search). The card is shared: the same values apply to AI Chatbot and AI Search. Most websites can keep the defaults. Click Configure on the card to open the settings drawer, change a value and click Save.
Training options explained
Chunk Size – default:1000
How many records are read from the database at a time while the content is prepared for training. When to change: lower it if the training stops because the server runs out of memory; raise it on strong servers to train large sites faster.
Batch Size – default: 100
How many items are sent to the AI provider in one request when the embeddings are created. Also used by the --batch-size option of the training command. When to change: lower it if your AI provider reports rate limits or time-outs; raise it to train faster when your provider allows larger requests.
Retention Days – default: 30
After how many days old Failed queue items are archived (CSV in var/log/t3cs) or deleted by the cleanup step. Completed (trained) items are always kept. When to change: raise it if you want to look into failed items for longer; lower it to keep the queue table small.
Embedding cache lifetime (seconds) – default: 2592000 (30 days)
How long the embeddings of visitor questions in search and chat are cached. A question that was asked before is answered without a new embedding call, which is faster and saves AI credits. When to change: lower it if you often switch the embedding provider or model; shorter values mean more AI calls.
Repeated blocks – default: 0.5 (allowed 0.2–0.9)
Removes text that repeats on many pages, such as promo banners, “shop the video” boxes, call-to-action boxes or legal notes. Such a block is kept once instead of on every page, so it does not push real content out of answers. A block counts as repeated when it appears on more than this share of the pages – across the whole site or within one URL folder (for example all /products/ pages). Empty or invalid values use 0.5.
- Lower (for example
0.3) – removes more repeated text. - Higher (for example
0.7) – keeps content that is shared on purpose, for example the same product card in several collections.
4 seconds
Follow-up questions (for example “and how much is it?”) are turned into a standalone search query with a short AI call. This setting is how long that call may take. If it takes longer, the question is searched as typed, so a slow AI service cannot block the answer. 0 uses the time limit of the AI provider. When to change: raise it if your AI provider is often slow and follow-up answers miss; lower it if answers must start faster.
AI Search (result list) – default: on
AI Search shows a numbered result list – one entry per page, with highlighted snippets – before the AI summary. The summary refers to these entries. Switch it off to show only the AI summary. It does not affect the chatbot.
All training settings and their defaults
All training settings and their defaults
Command line
The same commands are described for AI Chatbot.For developers: run training on the command line
For developers: run training on the command line
You can start the training yourself on the command line instead of waiting for the Scheduler, for example to test it or to train everything again. The commands come from the shared AI Chatbot/Search core and work the same for AI Chatbot and AI Search.Run them with Run the Scheduler task – Compound words – See all options with
vendor/bin/typo3 (Composer), php typo3/sysext/core/bin/typo3 (non-Composer) or ddev typo3 (DDEV).Training – nst3af:training [rootPageId]Syncs the data sources of a site, trains all Pending items (creates the embeddings) and cleans up old Failed items. rootPageId is the ID of the site root page. The aliases t3cs:training and aiuniverse:training run the same command.scheduler:run --task=<taskUid> -fRuns the site’s T3CS Training (Site N) task directly, with the options saved on the task. The task UID is shown in the Scheduler module.t3cs:keyword:compoundsAdds the parts of compound words (for example “Sonnen” and “schutz” from “Sonnenschutz”) to the keyword index of already trained content, so the keyword search also finds the single words. No AI calls, no new embeddings. Run it after a training on sites with compound words, for example German.vendor/bin/typo3 nst3af:training --help.For developers: run history cleanup on the command line
For developers: run history cleanup on the command line
t3af:history:cleanup [days] deletes AI Search and AI Chatbot usage history (visitor searches, conversations) and request traces older than days (default 90). It is separate from the training cleanup. The alias t3cs:history:cleanup runs the same command. You can also add it as a Scheduler task (Execute console commands).