2026 frontier model evaluations

GPT-5.6
Benchmark Comparison

Explore how the GPT-5.6 family compares across professional work, coding, science, computer use, cybersecurity, and more.

Benchmarks: 40
Models: 11
Categories: 11

01 / Explorer

Find a benchmark

40 of 40 charts

Models shown

OpenAI

Anthropic

Google

Professional

Agents' Last Exam

GPT-5.6 Sol52.7%

GPT-5.6 Terra50.4%

GPT-5.6 Luna50.3%

GPT-5.546.9%

Claude Opus 4.845.2%

Claude Fable 540.5%

Gemini 3.1 Pro Preview32.1%

Professional

GDPval-AA v2

Claude Fable 51,759.6 Elo

GPT-5.6 Sol1,747.8 Elo

Claude Opus 4.81,600.1 Elo

GPT-5.6 Terra1,593 Elo

GPT-5.6 Luna1,591.8 Elo

GPT-5.51,493.7 Elo

Gemini 3.5 Flash1,348.8 Elo

Gemini 3.1 Pro Preview962.3 Elo

Professional

Management Consulting Tasks (Internal)

GPT-5.6 Sol43.2%

GPT-5.6 Terra37.2%

Claude Fable 535.5%

GPT-5.6 Luna35.4%

Claude Opus 4.831.6%

GPT-5.531.3%

Gemini 3.1 Pro Preview13.2%

Professional

Big Finance Bench

GPT-5.6 Sol53%

GPT-5.6 Terra51%

GPT-5.549%

Claude Opus 4.844%

GPT-5.6 Luna36%

Professional

Artificial Analysis Intelligence Index v4.1

Claude Fable 559.9 Index score

GPT-5.6 Sol58.9 Index score

Claude Opus 4.855.7 Index score

GPT-5.6 Terra55 Index score

GPT-5.554.8 Index score

GPT-5.6 Luna51.2 Index score

Gemini 3.5 Flash50.2 Index score

Gemini 3.1 Pro Preview46.5 Index score

Coding

Artificial Analysis Coding Agent Index v1.1

GPT-5.6 Sol80 Index score

GPT-5.6 Terra77.4 Index score

Claude Fable 577.2 Index score

GPT-5.576.4 Index score

GPT-5.6 Luna74.6 Index score

Claude Opus 4.872.5 Index score

Gemini 3.1 Pro Preview42.7 Index score

Coding

SWE-Bench Pro

Claude Mythos 580.3%

Claude Fable 580%

Claude Mythos Preview77.8%

Claude Opus 4.869.2%

GPT-5.6 Sol64.6%

GPT-5.6 Terra63.4%

GPT-5.6 Luna62.7%

GPT-5.559.4%

Gemini 3.1 Pro Preview54.2%

Coding

DeepSWE v1.1

GPT-5.6 Sol72.7%

Claude Fable 569.7%

GPT-5.6 Terra69.6%

GPT-5.6 Luna67.2%

GPT-5.567%

Claude Opus 4.859%

Gemini 3.1 Pro Preview11.8%

Coding

Terminal-Bench 2.1

GPT-5.6 Sol Ultra91.9%

GPT-5.6 Sol88.8%

Claude Mythos 588%

GPT-5.6 Terra87.4%

GPT-5.585.6%

GPT-5.6 Luna84.7%

Claude Fable 583.1%

Claude Opus 4.878.9%

Gemini 3.1 Pro Preview70.7%

Science & health

GeneBench Pro

GPT-5.6 Sol28.7%

GPT-5.6 Terra23.3%

Claude Opus 4.816%

GPT-5.512%

GPT-5.6 Luna10.8%

Gemini 3.5 Flash8.14%

Gemini 3.1 Pro Preview3.1%

Science & health

LifeSciBench

GPT-5.6 Sol59.9%

GPT-5.6 Terra56%

Claude Opus 4.853.6%

GPT-5.6 Luna51.2%

GPT-5.550.4%

Science & health

MedChemBench (Internal)

GPT-5.6 Sol48.3%

GPT-5.535.5%

GPT-5.6 Terra35%

GPT-5.6 Luna30.4%

Science & health

HealthBench Professional⁶

Claude Fable 560.9%

GPT-5.6 Sol60.5%

GPT-5.6 Terra57.7%

GPT-5.6 Luna55.7%

Claude Opus 4.853%

GPT-5.549.5%

Computer use

OSWorld 2.0

GPT-5.6 Sol62.6%

Claude Opus 4.854.8%

GPT-5.6 Terra50.2%

GPT-5.547.5%

GPT-5.6 Luna45.6%

Computer use

BrowseComp

GPT-5.6 Sol Ultra92.2%

GPT-5.6 Sol90.4%

Claude Mythos 588%

Claude Mythos Preview87.9%

GPT-5.6 Terra87.5%

Gemini 3.1 Pro Preview85.9%

GPT-5.584.4%

Claude Opus 4.884.3%

GPT-5.6 Luna83.3%

Computer use

BenchCAD

GPT-5.6 Sol70.6%

GPT-5.6 Luna63.1%

GPT-5.6 Terra62.3%

GPT-5.544.4%

Claude Mythos 538.4%

Claude Mythos Preview35.5%

Claude Opus 4.827.3%

Computer use

BenchCAD (python tool)

GPT-5.6 Sol83.4%

GPT-5.6 Terra78.2%

GPT-5.6 Luna73.9%

Claude Mythos 565%

Claude Mythos Preview61%

GPT-5.555.8%

Claude Opus 4.851.8%

Cybersecurity

Capture-the-Flag Challenges

GPT-5.6 Sol96.7%

GPT-5.6 Terra91.8%

GPT-5.588.1%

GPT-5.6 Luna85.2%

Cybersecurity

SEC-Bench Pro

GPT-5.6 Sol Ultra74.3%

GPT-5.6 Sol71.2%

GPT-5.6 Terra57.7%

GPT-5.6 Luna48.9%

GPT-5.545.8%

Cybersecurity

CyberGym

GPT-5.6 Sol84.5%

Claude Mythos 583.8%

Claude Mythos Preview83%

GPT-5.6 Terra81.8%

GPT-5.581.8%

Claude Opus 4.878.1%

GPT-5.6 Luna77.9%

Cybersecurity

ExploitBench

Claude Mythos 578%

Claude Mythos Preview74.2%

GPT-5.6 Sol73.5%

GPT-5.6 Terra52.9%

GPT-5.547.9%

Claude Opus 4.840%

GPT-5.6 Luna33.2%

Cybersecurity

ExploitGym

GPT-5.6 Sol33.7%

GPT-5.6 Terra23.2%

GPT-5.515.1%

GPT-5.6 Luna12.4%

Self-improvement

Internal Research Debugging Evaluation

GPT-5.6 Sol68.3%

GPT-5.6 Terra67.8%

GPT-5.6 Luna50.8%

GPT-5.550%

Self-improvement

KernelGen 1P

GPT-5.6 Sol61.1%

GPT-5.6 Terra49.2%

GPT-5.529.3%

GPT-5.6 Luna22.4%

Self-improvement

NanoGPT

GPT-5.6 Terra14.5%

GPT-5.6 Sol9.69%

GPT-5.52.65%

GPT-5.6 Luna1.66%

Self-improvement

PostTrainBench Lite

GPT-5.6 Terra51.5%

GPT-5.6 Sol50.3%

GPT-5.538.8%

GPT-5.6 Luna29.6%

Self-improvement

RSI Index

GPT-5.6 Sol57.9%

GPT-5.6 Terra56.3%

GPT-5.6 Luna41.9%

GPT-5.541.7%

Multimodal

MMMU Pro (no tools)

GPT-5.6 Sol83%

GPT-5.581.2%

GPT-5.6 Terra80.7%

Gemini 3.1 Pro Preview80.5%

GPT-5.6 Luna78.4%

Multimodal

MMMU Pro (with tools)

GPT-5.6 Sol84.6%

GPT-5.583.2%

GPT-5.6 Terra82%

GPT-5.6 Luna79.5%

Multimodal

gdp.pdf

GPT-5.6 Sol30.7%

Claude Fable 529.8%

GPT-5.526%

GPT-5.6 Terra24.7%

GPT-5.6 Luna22.7%

Claude Opus 4.822.5%

Gemini 3.1 Pro Preview16.7%

Academic

GPQA Diamond

GPT-5.6 Sol94.6%

Claude Mythos Preview94.6%

Gemini 3.1 Pro Preview94.3%

Claude Mythos 594.1%

GPT-5.593.6%

GPT-5.6 Terra92.9%

Claude Fable 592.6%

GPT-5.6 Luna92.3%

Claude Opus 4.892%

Academic

FrontierMath Tier 1-3 (v2)

GPT-5.6 Sol89%

Claude Fable 587%

GPT-5.585.3%

GPT-5.6 Terra84.9%

Claude Opus 4.880%

GPT-5.6 Luna78.6%

Gemini 3.1 Pro Preview59.6%

Academic

FrontierMath Tier 4 (v2)

Claude Fable 587.8%

GPT-5.6 Sol83%

GPT-5.572.5%

GPT-5.6 Terra68.3%

GPT-5.6 Luna58.5%

Claude Opus 4.856.1%

Tool use

AutomationBench

GPT-5.6 Sol18.1%

Claude Fable 517.4%

Claude Opus 4.815.5%

GPT-5.6 Terra15.2%

GPT-5.6 Luna14.9%

Gemini 3.5 Flash14.5%

GPT-5.512.9%

Tool use

Toolathlon

Claude Mythos 561.7%

Claude Fable 561.7%

Claude Mythos Preview61.1%

Claude Opus 4.859.9%

GPT-5.6 Sol58%

GPT-5.555.6%

GPT-5.6 Luna53.4%

GPT-5.6 Terra53.1%

Gemini 3.1 Pro Preview48.8%

Long context

OpenAI MRCR v2 8-needle 256K-512K

GPT-5.6 Sol91.5%

GPT-5.6 Terra89.6%

GPT-5.581.5%

GPT-5.6 Luna41.3%

Long context

OpenAI MRCR v2 8-needle 512K-1M

GPT-5.574%

GPT-5.6 Sol73.8%

GPT-5.6 Terra72.5%

GPT-5.6 Luna41.3%

Long context

GraphWalks BFS 256k f1

Claude Mythos 591.1%

GPT-5.6 Sol90.7%

Claude Opus 4.885.9%

Claude Mythos Preview85.7%

GPT-5.6 Luna81.3%

GPT-5.6 Terra76.9%

GPT-5.573.7%

Long context

GraphWalks BFS 1mil f1

Claude Mythos 579.4%

GPT-5.6 Sol77.1%

Claude Mythos Preview74.3%

GPT-5.6 Terra71.2%

Claude Opus 4.868.1%

GPT-5.6 Luna51.2%

GPT-5.545.4%

Abstract reasoning

ARC-AGI-3⁷

GPT-5.6 Sol7.78%

Claude Opus 4.81.5%

GPT-5.6 Terra0.8%

GPT-5.50.43%

Gemini 3.1 Pro Preview0.42%

GPT-5.6 Luna0.18%

GPT-5.6Benchmark Comparison

Find a benchmark

GPT-5.6
Benchmark Comparison