GPT-5.3 Codex (xhigh)

February 5, 2026Last synced Apr 7, 2026, 4:00 PM

Blended Price

$4.81/M

Input Price

$1.75/M

Output Price

$14.00/M

Speed

81 tok/s

TTFT

63.33s

Benchmark Scores

Intelligence Index

External Benchmarks

AA-Omniscience Accuracy

How GPT-5.3 Codex (xhigh) Compares

Axes

X Axis

Y Axis

Bubble Size

Filters

Blended Price (USD)$0.00 – $30.0

AgMoBench Overall≥ 3.5

Providers

ai21-labsalibabaanthropicawsazurebaiducoheredeepseek

Show quadrants

anthropic

openai

google

meta

mistral

nvidia

kimi

xai

azure

deepseek

aws

cohere

baidu

ai21-labs

zai

reka-ai

xiaomi

alibaba

minimax

ibm

Bubble size = Context Window

16384.00

2000000.00

Compare with other models

GPQA Diamond

0.9 / 100

LiveCodeBench

— / 100

Terminal-Bench Hard

0.5 / 100

τ²-Bench

0.9 / 100

AA-Omniscience Indexaa_omniscience

9.9 / 100

AIME 2025 (Matrix)benchmark_matrix

94.0

GPQA Diamond (Matrix)benchmark_matrix

81.0

GSM8Kbenchmark_matrix

99.0

HumanEvalbenchmark_matrix

93.0

IFEvalbenchmark_matrix

92.0

LiveCodeBench (Matrix)benchmark_matrix

85.0

MATH-500 (Matrix)benchmark_matrix

96.0

MMLUbenchmark_matrix

94.0

OSWorldbenchmark_matrix

64.7

SWE-bench Probenchmark_matrix

56.8

Terminal-Bench 2.0benchmark_matrix

64.7

MMMUbenchmark_matrix

84.0 / 100

SWE-bench Verifiedbenchmark_matrix

56.8 / 100

AA-Omniscience Hallucination RatePredicted

97.9 / 100

Aider PolyglotPredicted

89.1 / 100

AIMEPredicted

0.3 / 30

AIME 2025Predicted

0.9 / 30

AlpacaEval 2.0 LCPredicted

21.9 / 100

AlpacaEval 2.0 RawPredicted

17.1 / 100

ARC-AGI-1Predicted

91.7 / 100

ARC-AGI-1 Cost per TaskPredicted

1.5

ARC-AGI-2Predicted

82.1 / 100

ARC-AGI-2 Cost per TaskPredicted

3.5

BFCL (Berkeley Function Calling)Predicted

56.8

BigCodeBench CompletePredicted

41.5 / 100

BigCodeBench InstructPredicted

32.6 / 100

AA Intelligence Index (Matrix)Predicted

66.4

AA Long Context Reasoning (Matrix)Predicted

76.7

AIME 2024Predicted

99.3

Arena-Hard AutoPredicted

83.3

BrowseCompPredicted

78.4

BRUMO 2025Predicted

99.6

CMIMC 2025Predicted

37.6

CritPtPredicted

14.5

HLE (Matrix)Predicted

19.9

HMMT Feb 2025Predicted

96.4

HMMT Nov 2025Predicted

96.0

IFBench (Matrix)Predicted

63.4

IMO 2025Predicted

42.9

MathArena Apex 2025Predicted

8.3

MMLU-Pro (Matrix)Predicted

86.7

MMMU-ProPredicted

80.8

MRCR v2Predicted

81.5

SimpleQAPredicted

67.6

SMT 2025Predicted

97.6

Tau-Bench Telecom (Matrix)Predicted

98.6

Terminal-Bench 1.0Predicted

48.4

USAMO 2025Predicted

33.3

Video-MMUPredicted

86.1

browsecompPredicted

78.1

Aider PolyglotPredicted

0.2

Arc Agi 2Predicted

0.1

BIG-Bench HardPredicted

3.0

CAD-EvalPredicted

2.4

Chess PuzzlesPredicted

0.3

CyBenchPredicted

0.8

DeepResearchBenchPredicted

0.5

FictionLiveBenchPredicted

1.0

GeoBenchPredicted

0.0

GSM8K (Epoch)Predicted

2.5

HellaSwagPredicted

0.0

Lech Mazur WritingPredicted

8.7

OTIS Mock AIME 2024–2025Predicted

1.3

SimpleQA Verified (Epoch)Predicted

0.6

The Agent CompanyPredicted

37.2

TriviaQAPredicted

20.8

WinoGrandePredicted

1.4

FrontierMathPredicted

35.2 / 100

GAIA Level 1Predicted

82.7

GAIA Level 2Predicted

68.2

GAIA Level 3Predicted

85.6

GAIAPredicted

68.9 / 100

LegalBenchPredicted

75.2 / 100

LiveCodeBenchPredicted

0.8 / 100

LongBench v2 EasyPredicted

77.8

LongBench v2 HardPredicted

72.5

LongBench v2Predicted

40.9 / 100

MATH-500Predicted

0.9 / 100

MathVistaPredicted

73.0 / 100

MedQA (USMLE)Predicted

96.0

MLE-benchPredicted

62.7 / 100

MMLU ProPredicted

0.9 / 100

MMTU Table UnderstandingPredicted

75.1 / 100

MT-BenchPredicted

8.3 / 10

NoLiMa (NIAH)Predicted

88.5 / 100

OCRBench v2Predicted

84.6 / 100

Open LLM AveragePredicted

12.5 / 100

Open LLM: BBHPredicted

39.0 / 100

Open LLM: GPQAPredicted

27.0 / 100

Open LLM: IFEvalPredicted

27.1 / 100

Open LLM: MATH Level 5Predicted

3.8 / 100

Open LLM: MMLU-PROPredicted

23.7 / 100

Open LLM: MUSRPredicted

38.5 / 100

SimpleBenchPredicted

67.4 / 100

simpleqaPredicted

59.8

SWE-bench LitePredicted

60.6 / 100

tau-bench RetailPredicted

90.4 / 100

Vectara Factual ConsistencyPredicted

86.1 / 100

Vectara Hallucination RatePredicted

13.9 / 100

WebArenaPredicted

12.3 / 100

WildBenchPredicted

37.9

BullshitBenchbullshitbench

24.0 / 100

Apex Agentsepoch_ai

3.3

METR Time Horizonsepoch_ai

54.7

Posttrainbenchepoch_ai

0.0

LiveBench Codinglivebench

78.2 / 100

LiveBench Data Analysislivebench

62.7 / 100

LiveBench Languagelivebench

80.1 / 100

LiveBench Mathlivebench

87.8 / 100

LiveBench Overalllivebench

72.8 / 100

LiveBench Reasoninglivebench

80.2 / 100

RE-Benchrebench

79.4

WeirdMLweirdml

79.3 / 100