#Why big files uploaded in knowledge base are cutted off in the content?

1 messages · Page 1 of 1 (latest)

signal nebulaBOT
#
AI-generated Answer for jvini6832

Short answer: large uploads get “cut off” because of enforced knowledge‑base size limits.

  • Non‑enterprise workspaces have a total knowledge‑base cap of 20MB or 300,000 characters. If an upload would exceed that, only part of the content can be stored/used, which can appear truncated. (elevenlabs.io)
  • If you attach a document to an agent without RAG, only up to ~300KB per item can be assigned to the agent. Enabling RAG raises the per‑item assignment cap to 10MB. (help.elevenlabs.io)
  • For RAG itself, there are additional per‑workspace indexing limits by plan (e.g., Free 1MB → Enterprise 1GB). Hitting these limits can prevent full indexing of large files. (elevenlabs.io)

What to do

  • Split big documents into smaller, focused files before uploading. (elevenlabs.io)
  • Enable RAG for the agent (to use larger per‑item sizes) and ensure your plan’s RAG indexing limits aren’t exceeded. (help.elevenlabs.io)
  • If you need higher limits, consider an enterprise plan. (elevenlabs.io)

Sources:
Knowledge base | ElevenLabs Documentation
What can I create with ElevenLabs Agents (formerly Conversational AI) – ElevenLabs
Retrieval-Augmented Generation | ElevenLabs Documentation
Knowledge base | ElevenLabs Documentation
What can I create with ElevenLabs Agents (formerly Conversational AI) – ElevenLabs
Knowledge base | ElevenLabs Documentation

sturdy hatch
#

That's incorrect. I'm using the entreprise plano and I have simple files with 2MB that are losing caracteres like "#" and "---"