Skip to content

AndesGPT

A personalised large model and agent built on device-cloud collaboration

Visit official site

What AndesGPT is

The model is described as a personalised large model and agent released by OPPO, built on a device-cloud collaborative architecture with several parameter scales. It supports dialogue enhancement, personalisation and device-cloud collaboration as technical features.

What you can do with it

  • Use dialogue enhancement in conversation
  • Rely on personalisation for a dedicated experience
  • Work through device-cloud collaboration
  • Choose among several parameter-scale model specifications

Example uses

  • Rich agents in ColorOS 14 offer different scene experiences
  • Multi-turn dialogue receives fast system responses
  • Music generation played a role in a released single

Model notes

  • A joint intelligent computing lab is set up with the University of Science and Technology of China
  • An open agent framework supports efficient incubation, hosting and application
  • The architecture is described as device-cloud collaborative
  • Model specifications come in several different parameter scales

Pros & cons

✓ What we like

  • Multiple parameter scales are offered as model specifications
  • An open agent framework supports efficient incubation and hosting
  • Rich agents are provided in ColorOS 14 for different scenes

! What to watch out for

  • The listed features centre on dialogue enhancement, personalisation and device-cloud collaboration
  • Capabilities depend on models offered at several parameter scales
  • The open agent framework and joint lab are described as part of forward-looking plans

FAQ

Who released it?

The material describes it as a personalised large model and agent released by OPPO.

What technical features are named?

Dialogue enhancement, personalisation and device-cloud collaboration are listed.

Where are the agents used?

The material names ColorOS 14, multi-turn dialogue and a released single for music generation.

Last reviewed: 2026-11-15

More LLM API platform tools

View all →

How we review