AI vendors and your data

146 vendors · 427 answers

Secure model inference changed Baseten, Oct 26, 2026

Baseten · Oct 26, 2026 · 1 added, 1 removed · seen by the weekly check

  1. Added: Model inputs and outputs: Baseten stores async inference inputs while requests wait in the queue and while models execute them. Requests can remain queued for up to 72 hours. Baseten doesn't store outputs.
  2. Removed: Model inputs and outputs: Baseten temporarily stores async inference inputs until processing completes. It does not store outputs.
About this change
Page
docs.baseten.co/observability/security
Kind
Documentation
Text hash
bc1ee9c06d19 to 6923f1e6041f
Dated by
the weekly check that first read the new text