Model

GPT-5.6 Sol

CLOSED

OpenAI · GPT-5.6 family · released Jul 9, 2026

Flagship (Sol) tier of the three-tier GPT-5.6 family, billed as OpenAI's best coding model yet (Terminal-Bench 2.1 SOTA); limited-preview June 26 under US govt review, GA July 9, 2026. Sibling tiers GPT-5.6 Terra and GPT-5.6 Luna are tracked as their own corpus models in the same family (OpenAI: 'durable capability tiers that can advance on their own cadence'). Base price is unchanged at $5/$30 per MTok in the July 30, 2026 round of cuts that took Terra to $2/$12 and Luna to $0.20/$1.20; Sol instead gained a Fast mode in the API — up to 2.5x the speed at twice the price ($10/$60) with the same intelligence, so it is a serving option rather than a separate tier and is not filed as its own model row.

ReasoningCodingVisionFunction callingTool useAgentic

2759.9

Elo · rank #18

Parameters

Undisclosed

Active params

—

Context

1.05M tokens

Architecture

Undisclosed architecture (presumed dense transformer); flagship tier of the three durable capability tiers Sol/Terra/Luna, each tracked as its own corpus model (like Claude Opus/Sonnet/Haiku)

License

Proprietary

Languages

—

API price (in/out)

$5 / $30

Modalities

text · vision

Benchmark results

Bar shows position within the tracked field; marker = field best

AA-BriefcaseAgents1505#5

best: Claude Opus 5 · 1720

Agents' Last ExamAgents52.7%#2

best: GPT-6 Astra · 59.3%

ARC-AGI-1Reasoning96.5%#9

best: Claude Fable 5 · 98.5%

ARC-AGI-2Reasoning92.5%#4

best: GPT-6 Astra · 95.0%

ARC-AGI-3Reasoning7.8%#6

best: GPT-6 Astra · 62.7%

Arena EloHuman preference1485#15

best: Muse Wonder · 1513

AutomationBenchAgents18.1%#18

best: MiniMax M3.1 Flash · 62.8%

BenchCAD (Vision2Code)Vision83.3%#4

best: GPT-6 Astra · 95.9%

BrowseCompAgents90.4%#4

best: Kimi K3.1 Pro · 92.1%

CursorBenchCoding67.2%#5

best: Claude Fable 5.1 · 73.4%

CyberGymCoding84.5%#7

best: MiMo-V2.6-Flash · 95.1%

DeepSearchQAAgents93.0%#1

best: this model · 93.0%

DeepSWECoding72.7%#10

best: Muse Wonder Full · 79.0%

ExploitBenchAgents76.5%#5

best: GPT-6 Astra · 100.0%

FrontierBenchAgents34.4%#2

best: Claude Opus 5 · 43.3%

FrontierCode (Extended)Coding60.6%#7

best: Claude Opus 5.5 · 65.3%

FrontierCodeCoding47.5%#8

best: Claude Opus 5.5 · 54.6%

FrontierMath (Tiers 1-3)Math89.0%#1

best: this model · 89.0%

FrontierMath (Tier 4)Math83.0%#1

best: this model · 83.0%

GDP.PDFVision40.0%#1

best: this model · 40.0%

GDPval-AAAgents1747.8#8

best: Claude Fable 5 · 1932

GPQA DiamondReasoning94.6%#2

best: GPT-6 Astra · 96.0%

Harvey Legal Agent BenchmarkAgents2.5%#34

best: Muse Spark 1.2 · 25.4%

HLE-VerifiedKnowledge54.5%#2

best: Gemini 3.8 Flash · 54.9%

Humanity's Last ExamReasoning47.2%#17

best: Kimi K3.1 Pro · 70.0%

IFBenchReasoning72.7%#26

best: MiniMax M3 · 83.0%

JobBenchAgents45.4%#12

best: Muse Wonder Full · 75.3%

LVBenchVision82.1%#4

best: Gemini 4 Argon · 91.7%

MMMU-ProVision83.0%#4

best: Claude Opus 4.7 · 85.5%

MRCR v2 8-needle (512K-1M)Reasoning73.8%#3

best: Muse Spark 1.3 · 98.1%

MRCR v2 8-needle (256K-512K)Reasoning91.5%#3

best: GPT-6 Astra · 100.0%

OSWorld 2.0Agents62.6%#10

best: Claude Opus 5.5 · 81.8%

ScreenSpot-ProVision76.9%#3

best: GPT-6 Astra · 92.7%

SRE-BenchCoding55.9%#2

best: GPT-6 Astra · 88.0%

SWE-Atlas Codebase QnACoding53.5%#5

best: Step 5 Preview · 63.6%

SWE-bench ProCoding64.6%#8

best: Claude Opus 5.5 · 89.9%

SWE-bench VerifiedCoding96.2%#2

best: Claude Opus 5 · 97.0%

SWE-MarathonCoding42.5%#10

best: Muse Wonder Full · 57.5%

τ³-BankingAgents33.0%#2

best: Step 5 Preview · 42.5%

Terminal-Bench 2.0Coding88.8%#4

best: DeepSeek-V4.1-Flash · 90.6%

Terminal-Bench 3.0Agents34.6%#1

best: this model · 34.6%

Terminal-Bench 4.0Agents37.9%#17

best: Claude Sonnet 5.5 · 70.6%

Terminal-Bench Science 0.1Agents20.0%#11

best: GPT-6 Astra · 62.9%

Toolathlon (Verified)Agents74.9%#5

best: GLM-5.3-Flash · 78.4%

Vals Finance AgentAgents53.8%#27

best: Gemini 4 Argon · 65.4%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

$5

Output / M tok

$30

GPT-5.6 family

Elo progression across releases

GPT-5.6 LunaJul 20262473.6 GPT-5.6 SolJul 20262759.9 GPT-5.6 TerraJul 20262584.3 GPT-5.6 CyberAug 20261000.0

Resources

DOCSdevelopers.openai.com/api/docs/models/gpt-5.6-sol APIAvailable via OpenAI API

API price $5/$30 · each benchmark row carries its own source badge (see methodology)