GPT-5.6 Terra

The GPT-5.6 balance point for intelligence, speed and cost across scaled professional workloads.

Context1.05M

ModalityText · Image

AccessAPI

ReasoningAdaptive

Where it fits.

A pragmatic production default when Sol is excessive and Luna leaves too little capability margin. Terra keeps the same 1.05M context, 128K output ceiling, image input and production tool set as Sol while targeting the intelligence-cost middle tier.

Core capabilities

1.05M context and 128K maximum output
Text and image input with multilingual support
Configurable reasoning from none to max
Structured outputs and the full Responses API tool suite

Best-fit workflows

Enterprise knowledge assistants
Scaled content and analysis
General-purpose agents

04 / OPERATIONS

From evaluation
to production.

Deployment notes

A strong default for enterprise assistants and general agents. Route only the hardest cases to Sol and high-volume deterministic subtasks to Luna; keep shared prompts and evals so routing decisions remain measurable.

Watch before choosing

Measure end-to-end cost using real prompt length, tool calls and reasoning settings. Its lower unit price does not remove long-context surcharges or tool-call costs. Compare total workflow completion cost, not only token price.

Family & versions

GPT-5.6 SolFlagship
GPT-5.6 TerraBalanced
GPT-5.6 LunaEfficient

Specifications and positioning checked against provider documentation.