Google Cloud page describes Gemini distillation service, but its release status is unclear
An archived Google document outlines an allowlist-only service, but a production ban, conflicting model IDs and unfinished tooling leave its release status unclear.
By Ryan Merket · Published
Primary source: Google Cloud Documentation via Wayback Machine
Why it matters
Managed distillation could let AI companies serve specialized reasoning workloads at Flash-tier speed and cost without creating labeled answers, while tying training and deployment to Google Cloud.
An archived Google Cloud documentation page describes an allowlisted service that would use Gemini 3.1 Pro outputs and reasoning patterns to train custom Gemini 2.5 Flash models. The page says it was updated July 23rd, but its contents do not establish that Google has formally launched the product.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.