/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents system to pick the right AI size for each question

Google describes a method for choosing between different artificial intelligence models based on how complex a user's request is. The system tests various models to see which ones handle specific types of questions well, then trains a smaller program to automatically select the most efficient model for future requests.

Why it mattersThis approach could make AI responses faster and cheaper by avoiding the use of massive, slow models for simple tasks.

In the filing’s own words

Implementations disclose selecting, in response to receiving a request and from among multiple candidate generative models (e.g., multiple candidate large language models (LLMs)) with differing computational efficiencies, a particular generative model to utilize in generating a response to the request. Those implementations reduce latency and/or conserve computational resource(s) through selection, for various requests, of a more computationally efficient generative model for utilization in lieu of a less computationally efficient generative model. Further, those implementations seek to achieve such benefits, through utilization of more computationally efficient generative models, while also still selectively utilizing less computationally efficient generative models for certain requests to mitigate occurrences of a generated response being inaccurate and/or under-specified. This, in turn, can mitigate occurrences of computational and/or network inefficiencies that result from a user issuing a follow-up request to cure the inaccuracies and/or under-specification of a generated response.

What exactly they claim (claim 1)

1. A method implemented by one or more processors, the method comprising: identifying request features for a request; for each of a plurality of generative models: causing the request to be processed, using the generative model, to generate corresponding generative model output; and generating a corresponding measure, for the generative model, based on the corresponding generative model output; determining, based on the corresponding measures, ground truth classification labels, each of the ground truth classification labels being for a corresponding one of the generative models; generating a training instance that includes: training instance input that includes the request features, and training instance output that includes the ground truth classification labels; and causing training of a machine learning model using the training instance.

Facts

Company
Google (GOOGLE LLC)
Official title
DYNAMIC SELECTION FROM AMONG MULTIPLE CANDIDATE GENERATIVE MODELS WITH DIFFERING COMPUTATIONAL EFFICIENCIES
Publication number
US20260288789A1
Application number
19684709
Filed
May 21, 2026
Published
Sep 24, 2026
Inventors
19
Claims
20

Classified under

G06F 16/3329 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.