Models

Amp uses the best model for each task: leading generalist foundation models for complex reasoning and planning, and smaller specialized models for fast, accurate responses in specific domains.

Agent Modes

Combination of System Prompt + Tools + Model

Deep

The most capable coding mode, with deep reasoning

GPT-5.5

Smart

Strong intelligence for any task

Claude Opus 4.8

Rush

Fast, low-token work for small, well-defined tasks

GPT-5.5

Subagents

Review

Bug identification & code review assistance

Gemini 3.1 Pro

Fast, accurate codebase retrieval

Gemini 3 Flash

Oracle

Complex reasoning & planning on code

GPT-5.5

Librarian

Large-scale retrieval & research on external code

Claude Sonnet 4.6

Read Thread

Reading & summarizing other Amp threads

Gemini 3.5 Flash

System Models

View Media

Analysis of images, PDFs, audio, and video

Gemini 3 Flash

Painter

Image generation and editing

GPT Image 2

Handoff

Fallback context analysis for task continuation

Gemini 3 Flash

Titling

Fast title generation for threads

Claude Haiku 4.5

Model Cards

Evals, caveats, and rollout notes behind Amp’s model choices.

GPT 5.5

GPT-5.5 now powers deep, rush, and Oracle. It is more capable, but can cost more.

Read card

See the Amp Security Reference for more details on model use and inference providers.