GPT-5.6 Sol

Jul 2026
GPT-5.6 Sol
Compare in
on
$0.26per task?
$4 / $20per 1M in / outcache $0.40
77 tok/sfirst token in 2.1 s?
1.05Mcontext

Rankingat low effort?

Full ranking →
1Fable 5.167.22Opus 5.566.23Fable 565.44GPT-6 Astra62.65GPT-6.1 Sol62.1
···
7GPT-5.5588GPT-5.6 Sol57.29MiniMax M3.1 Flash56.2

Strong at

S
Knowledge#27 of 318Vision#12 of 124
A
Reasoning#48 of 334Science#35 of 312Long context#67 of 273
B
Coding#53 of 132Agents & work tasks#46 of 175Tool calling#85 of 242Instructions#66 of 245

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperGPT-6.1 Solmedium+6%-18%$0.21Remove unsupported parameters+5OpenDeepSeek V4 Pro 0813max-14%+159%$0.67Context: 1.05M → 1M+6CheaperMiniMax M3.1 Flashthinkingsame-16%$0.22Context: 1.05M → 1M+1CheaperGPT-6 Solmedium-7%-5%$0.25CheaperGemini 3 Flash Previewthinking-7%-86%$0.50 / $3Context: 1.05M → 1M

History

3 versions
5456586062Oct 26Release dateAA Omniscience · Accuracy5.666.1
5456586062Oct 26Release dateAA Omniscience · Accuracy5.666.1

Benchmarks

in Chat & Knowledge
AA Intelligence Index 4.3.233.5#49
AA Omniscience Index18.9#27
AA Omniscience · Accuracy57.2#8
AA Omniscience · Hallucination Rate89.4#242
Humanity's Last Exam39.4#48
GPQA Diamond89.8#35
AA-LCR (long context) 1.178#67
IFBench (instructions)66.5#66
CritPt (physics)14.9#45
GDP.pdf · AA all-pass21#22
MLCR-AA12.8#36
ARC-AGI-1 · public evaluation84.5#34
ARC-AGI-1 · semi-private74.5#37
ARC-AGI-2 · public evaluation38.5#32
ARC-AGI-2 · semi-private42.5#33
ARC-AGI-3 · semi-private0.3#14
GPQA Diamond · Epoch89.9#33
OTIS Mock AIME 2024 202595.6#31
Chess Puzzles31#41
Mystery Game Puzzles26#27
HLE fulllab44.5
SEC-Bench Prolab74.3
ExploitGymlab33.7
HealthBench Professional length-adjustedlab60.5
HealthBench length-adjustedlab57
HealthBench Hard length-adjustedlab33.1
HealthBench Consensus length-adjustedlab95.5

FAQ

How much does GPT-5.6 Sol cost?›

$4 per 1M input tokens and $20 per 1M output tokens; cached input $0.40 per 1M. A typical benchmark task costs about $0.26 at low effort.

Is GPT-5.6 Sol free to use?›

There is no free API; the cheapest paid price is $4 per 1M input tokens.

How good is GPT-5.6 Sol, and what is it best at?›

#48 of 334 in Chat & Knowledge, #53 of 132 in Coding, #41 of 133 in Agents & Tool Use, #12 of 195 in Writing & Language, #12 of 124 in Vision & Documents. Strongest at: Reasoning, Science, Long context, Knowledge, Vision.

Is GPT-5.6 Sol better than GPT-4?›

On AA Omniscience · Accuracy, GPT-5.6 Sol scores 59.4 vs 21 — higher.

How can I use GPT-5.6 Sol?›

Through the API of Amazon Bedrock, Azure, OpenAI.

What is the context window of GPT-5.6 Sol?›

1.05M tokens, with up to 128K output tokens.

When was GPT-5.6 Sol released? Is there a newer version?›

Released 2026-07-09. It has been superseded by GPT-6 Sol.

Is GPT-5.6 Sol open source?›

No, GPT-5.6 Sol is a closed model available through an API. The closest open-weights option here is DeepSeek V4 Pro 0813.

Can GPT-5.6 Sol read images?›

It accepts text and images, PDF files as input.

What is GPT-5.6 Sol?›

GPT-5.6 Sol is a closed language model from OpenAI, released Jul 2026, the successor to GPT-4.

What are the best alternatives to GPT-5.6 Sol?›

GPT-6.1 Sol (cheaper), DeepSeek V4 Pro 0813 (open weights).