← Models

GPT-5.6 Terra

OpenAI · GPT-5.6

Balanced GPT-5.6 model for production coding, knowledge work, agents, and long-context workflows at half Sol's token price.

Part of GPT family · Other versions: GPT-5.6 Luna , GPT-5.6 Sol , GPT-5.5 Pro , GPT-5.5 , GPT-Rosalind , GPT-5.4 mini , GPT-5.4 nano , GPT-5.4 Pro , GPT-5.4 , GPT-5.3 , GPT-5.3-Codex , GPT-5.2-Codex , GPT-5.2 Pro , GPT-5.2 , GPT-5-Codex , GPT-5 mini , GPT-5 nano , GPT-5 , GPT-4.1 , GPT-4o mini Transcribe , GPT-4o mini TTS , GPT-4o Transcribe , Computer Use Preview , GPT-4o mini , GPT-4o
Type
language
Context
1M tokens
Max Output
128K tokens
Status
current
Input
$2.5/1M tok
Output
$15/1M tok
API Access
Yes
License
proprietary
reasoning coding agentic tool-use computer-use long-context balanced production
Released July 2026 · Updated July 10, 2026

Overview

Freshness note: Model capabilities, limits, pricing, and availability can change quickly. This profile is a point-in-time snapshot last verified on July 10, 2026.

GPT-5.6 Terra is OpenAI’s balanced model for teams that want the GPT-5.6 generation without paying flagship Sol rates on every request. OpenAI describes it as the tier that balances intelligence and cost, roughly corresponding to the mini position used in earlier GPT-5 families even though its documented context and output ceilings match Sol.

Terra became generally available with the rest of GPT-5.6 on July 9, 2026. It is the most natural first evaluation target for broad production workloads that have outgrown older mini models but do not consistently need the highest reasoning ceiling.

Capabilities

Terra covers coding, knowledge work, tool use, computer use, document analysis, design tasks, and long-running agent workflows. It shares the GPT-5.6 generation’s support for Programmatic Tool Calling, persisted reasoning, explicit prompt caching, and the Responses API multi-agent beta.

Reasoning effort supports none, low, medium, high, xhigh, and max. OpenAI’s migration guidance recommends starting from the reasoning setting already used with GPT-5.5 or GPT-5.4, then testing the same setting and one level lower because GPT-5.6 may reach the target quality with fewer output tokens.

Technical Details

OpenAI’s API docs list:

  • Model ID: gpt-5.6-terra
  • Context window: 1,050,000 tokens
  • Maximum output: 128,000 tokens
  • Knowledge cutoff: February 16, 2026
  • Input: text and images
  • Output: text

Streaming, function calling, and structured outputs are supported. Terra also participates in the GPT-5.6 Responses tool ecosystem for search, files, code execution, shell work, patching, skills, computer use, MCP, and tool search. Fine-tuning is not currently supported.

Pro mode is configured with reasoning.mode: "pro" on Terra itself when a workflow needs more model work. There is no separate Terra Pro model slug in the current API catalog.

Pricing & Access

Published standard API pricing per 1M tokens is:

  • Input: $2.50
  • Cached input: $0.25
  • Output: $15.00

That is half Sol’s standard token price. Prompts above 272K input tokens use the GPT-5.6 long-context multiplier of 2x input and 1.5x output for the full request. Explicit cache writes cost 1.25x the uncached input rate; cached reads retain the lower published rate.

Terra is available through the OpenAI API, ChatGPT Work, and Codex, with product access varying by plan and workspace.

Best Use Cases

Choose Terra for production coding agents, document-heavy assistants, research synthesis, computer-use workflows, internal operations, and professional drafting where quality matters but Sol-level spend is difficult to justify at scale.

It is also a strong subagent or routing tier: use Luna for simple high-volume work, Terra for the broad middle, and Sol only when evaluation shows a meaningful quality gain.

Comparisons

  • GPT-5.6 Sol (OpenAI): Higher-capability flagship at twice Terra’s standard token price.
  • GPT-5.6 Luna (OpenAI): Cheaper route for high-volume and narrower workloads.
  • GPT-5.5 (OpenAI): Older frontier route with the same headline standard token price as Sol.
  • GPT-5.4 mini (OpenAI): Lower-cost earlier-generation mini model for workloads that do not need the GPT-5.6 feature set.