GPT-5.6 Terra

Name: GPT-5.6 Terra
Author: OpenAI

Multimodal

OpenAI

GPT-5.6 Terra is the balanced tier of OpenAI's GPT-5.6 family, designed for workloads that balance intelligence and cost. It delivers performance competitive with GPT-5.5 at roughly half the price, supports max reasoning effort, and has a 1.05M-token context window.

Key Specifications

Parameters

Context

1.1M

Release Date

July 9, 2026

Average Score

59.8%

API Documentation Research Paper Results Blog

Timeline

Key dates in the model's history

Announcement

July 9, 2026

Today

August 27, 2026

Technical Specifications

Parameters

Training Tokens

Knowledge Cutoff

February 16, 2026

Family

Capabilities

MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)

$2.00

Output (per 1M tokens)

$12.00

Max Input Tokens

1.1M

Max Output Tokens

128.0K

Supported Features

Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Reasoning

Logical reasoning and analysis

GPQA

GPQA Diamond. Max reasoning effort. • Self-reported

92.9%

Other Tests

Specialized benchmarks

Agents' Last Exam

Agents' Last Exam. Max reasoning effort. • Self-reported

50.4%

ARC-AGI-3

ARC-AGI-3. Max reasoning effort. • Self-reported

0.8%

Artificial Analysis

Artificial Analysis Intelligence Index v4.1 (max). Score 55. • Self-reported

55.0%

AutomationBench

AutomationBench. Max reasoning effort. • Self-reported

15.2%

BenchCAD

BenchCAD (no tools). Max reasoning effort. • Self-reported

62.3%

BenchCAD (with Python tool)

BenchCAD with Python tool. Max reasoning effort. • Self-reported

78.2%

Big Finance Bench

Big Finance Bench. Max reasoning effort. • Self-reported

51.0%

BrowseComp

BrowseComp. Max reasoning effort. • Self-reported

87.5%

Capture-the-Flag Challenges (Internal)

Capture-the-Flag Challenges (Internal). Max reasoning effort. • Self-reported

91.8%

Connectors

Connectors production benchmark pass rate. • Self-reported

100.0%

DeepSWE

DeepSWE v1.1. Max reasoning effort. • Self-reported

69.6%

DeepSWE 1.1

DeepSWE v1.1 leaderboard, mini-swe-agent harness, max effort; Pass@1 70% ± 3%. • Self-reported

70.0%

ExploitBench

ExploitBench. Max reasoning effort. • Self-reported

52.9%

ExploitGym

ExploitGym, six-hour cap. Max reasoning effort. • Self-reported

23.2%

FrontierCode 1.1

FrontierCode 1.1 current leaderboard; mergeability score at max effort: 41.3%. • Self-reported

41.3%

FrontierMath

FrontierMath Tier 1-3 (v2). Max reasoning effort. • Self-reported

84.9%

FrontierMath Tier 4 (v2)

FrontierMath Tier 4 (v2). Max reasoning effort. • Self-reported

68.3%

GDP.pdf

gdp.pdf. Max reasoning effort. • Self-reported

24.7%

GeneBench-Pro

GeneBench Pro. Max reasoning effort. • Self-reported

23.3%

Graphwalks BFS >128k

GraphWalks BFS 256k f1. Max reasoning effort. • Self-reported

76.9%

Graphwalks BFS 1M

GraphWalks BFS 1M f1. Max reasoning effort. • Self-reported

71.2%

HealthBench

HealthBench, length-adjusted score (unadjusted 58.7). • Self-reported

57.0%

HealthBench Consensus

HealthBench Consensus, length-adjusted score (unadjusted 95.2). • Self-reported

95.1%

HealthBench Hard

HealthBench Hard, length-adjusted score (unadjusted 34.3). • Self-reported

32.7%

HealthBench Professional

HealthBench Professional, length-adjusted score (unadjusted 62.4). • Self-reported

57.7%

Internal Research Debugging Evaluation

Internal Research Debugging Evaluation. Max reasoning effort. • Self-reported

67.8%

KernelGen 1P

KernelGen 1P. Max reasoning effort. • Self-reported

49.2%

LifeSciBench

LifeSciBench. Max reasoning effort. • Self-reported

56.0%

Management Consulting Tasks (Internal)

Management Consulting Tasks (Internal). Max reasoning effort. • Self-reported

37.2%

MedChemBench (Internal)

MedChemBench (Internal). Max reasoning effort. • Self-reported

35.0%

MMMU-Pro

MMMU Pro (no tools). Max reasoning effort. • Self-reported

80.7%

MMMU-Pro (with tools)

MMMU Pro (with tools). Max reasoning effort. • Self-reported

82.0%

MRCR v2 (8-needle)

OpenAI MRCR v2 8-needle, 256K-512K. Max reasoning effort. • Self-reported

89.6%

MRCR v2 (8-needle, 512K-1M)

OpenAI MRCR v2 8-needle, 512K-1M. Max reasoning effort. • Self-reported

72.5%

NanoGPT

NanoGPT. Max reasoning effort. • Self-reported

14.5%

OSWorld 2.0

OSWorld 2.0 binary completion. Max reasoning effort. • Self-reported

50.2%

PostTrainBench Lite

PostTrainBench Lite mean reward. Max reasoning effort. • Self-reported

51.5%

RSI Index

RSI Index (aggregate self-improvement). Max reasoning effort. • Self-reported

56.3%

Search and Function-Calling

Search and Function-Calling production benchmark pass rate. • Self-reported

94.6%

SEC-bench Pro

SEC-Bench Pro. Max reasoning effort. • Self-reported

57.7%

SWE-Bench Pro

SWE-Bench Pro. Max reasoning effort. • Self-reported

63.4%

Terminal-Bench 2.1

Terminal-Bench 2.1. Max reasoning effort. • Self-reported

87.4%

Toolathlon

Toolathlon. Max reasoning effort. • Self-reported

53.1%

License & Metadata

License

proprietary

Announcement Date

July 9, 2026

Last Updated

August 27, 2026

Compare GPT-5.6 Terra

All comparisons

vs GPT-5.6 Sol vs GPT-5.6 Luna vs GPT-5.1 Thinking vs GPT-5 High vs GPT-5.4 vs GPT-5.1 High vs GPT-5.1 Codex High vs GPT-5.2

Similar Models

All Models

GPT-5.5

OpenAI

Best score:1.0 (TAU)

Released:Apr 2026

Price:$5.00/1M tokens

GPT-5.1 Instant

OpenAI

Best score:1.0 (TAU)

Released:Nov 2025

Price:$0.30/1M tokens

GPT-5.6 Sol

OpenAI

Best score:0.9 (GPQA)

Released:Jul 2026

Price:$5.00/1M tokens

GPT-5.6 Luna

OpenAI

Best score:0.9 (GPQA)

Released:Jul 2026

Price:$0.20/1M tokens

GPT-5.2

OpenAI

Best score:1.0 (TAU)

Released:Dec 2025

Price:$1.50/1M tokens

GPT-5.4

OpenAI

Best score:1.0 (TAU)

Released:Mar 2026

Price:$2.50/1M tokens

GPT-5.1 Thinking

OpenAI

Best score:1.0 (TAU)

Released:Nov 2025

Price:$3.00/1M tokens

GPT-5 Medium

OpenAI

Best score:0.9 (GPQA)

Released:Aug 2025

Price:$0.75/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.