GPT-5 mini (minimal)

August 7, 2025Last synced Apr 7, 2026, 4:00 PM

Blended Price

$0.69/M

Input Price

$0.25/M

Output Price

$2.00/M

Speed

86 tok/s

TTFT

0.83s

Benchmark Scores

Intelligence Index

External Benchmarks

ARC-AGI-1

How GPT-5 mini (minimal) 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

0.7 / 100

LiveCodeBench

0.5 / 100

Terminal-Bench Hard

0.1 / 100

τ²-Bench

0.3 / 100

ARC-AGI-1 Cost per Taskarc_agi

0.0

ARC-AGI-2arc_agi

1.7 / 100

ARC-AGI-2 Cost per Taskarc_agi

0.0

AA-Omniscience AccuracyPredicted

40.7 / 100

AA-Omniscience Hallucination RatePredicted

95.7 / 100

Aider PolyglotPredicted

60.2 / 100

AIMEPredicted

0.3 / 30

AlpacaEval 2.0 LCPredicted

11.7 / 100

AlpacaEval 2.0 RawPredicted

9.1 / 100

BigCodeBench CompletePredicted

54.8 / 100

BigCodeBench InstructPredicted

44.3 / 100

AA Intelligence Index (Matrix)Predicted

67.5

AA Long Context Reasoning (Matrix)Predicted

73.3

AIME 2024Predicted

92.5

AIME 2025 (Matrix)Predicted

96.2

Arena-Hard AutoPredicted

72.9

BrowseCompPredicted

68.8

BRUMO 2025Predicted

96.2

CMIMC 2025Predicted

84.8

GPQA Diamond (Matrix)Predicted

79.3

HLE (Matrix)Predicted

24.1

HMMT Feb 2025Predicted

79.9

HMMT Nov 2025Predicted

77.4

HumanEvalPredicted

90.6

IFBench (Matrix)Predicted

51.8

IFEvalPredicted

88.0

IMO 2025Predicted

25.3

LiveCodeBench (Matrix)Predicted

70.4

MATH-500 (Matrix)Predicted

96.8

MathArena Apex 2025Predicted

2.3

MMLU-Pro (Matrix)Predicted

81.5

MMMU-ProPredicted

80.6

MRCR v2Predicted

79.1

OSWorldPredicted

43.7

SimpleQAPredicted

39.8

SMT 2025Predicted

87.4

SWE-bench ProPredicted

42.8

Tau-Bench Telecom (Matrix)Predicted

49.3

Terminal-Bench 2.0Predicted

36.7

Terminal-Bench 1.0Predicted

27.3

USAMO 2025Predicted

22.0

Video-MMUPredicted

86.1

browsecompPredicted

67.2

BullshitBenchPredicted

26.4 / 100

Aider PolyglotPredicted

0.0

Apex AgentsPredicted

2.6

BIG-Bench HardPredicted

3.0

CAD-EvalPredicted

4.0

Chess PuzzlesPredicted

0.2

CyBenchPredicted

0.4

DeepResearchBenchPredicted

0.4

GeoBenchPredicted

0.0

GSM8K (Epoch)Predicted

3.8

HellaSwagPredicted

0.1

Lech Mazur WritingPredicted

7.9

METR Time HorizonsPredicted

4.4

PosttrainbenchPredicted

0.0

The Agent CompanyPredicted

6.6

TriviaQAPredicted

7.7

WinoGrandePredicted

0.9

FrontierMathPredicted

20.4 / 100

GAIA Level 1Predicted

61.2

GAIA Level 2Predicted

37.5

GAIA Level 3Predicted

11.3

GAIAPredicted

44.1 / 100

LegalBenchPredicted

82.5 / 100

LongBench v2 EasyPredicted

70.9

LongBench v2 HardPredicted

42.9

LongBench v2Predicted

39.0 / 100

MATH-500Predicted

0.9 / 100

MathVistaPredicted

65.3 / 100

MLE-benchPredicted

38.0 / 100

MMMUPredicted

76.2 / 100

MMTU Table UnderstandingPredicted

63.8 / 100

MT-BenchPredicted

7.9 / 10

NoLiMa (NIAH)Predicted

93.0 / 100

OCRBench v2Predicted

82.8 / 100

Open LLM AveragePredicted

12.3 / 100

Open LLM: BBHPredicted

39.4 / 100

Open LLM: GPQAPredicted

27.9 / 100

Open LLM: IFEvalPredicted

29.2 / 100

Open LLM: MATH Level 5Predicted

47.1 / 100

Open LLM: MMLU-PROPredicted

22.0 / 100

Open LLM: MUSRPredicted

38.4 / 100

RE-BenchPredicted

78.9

SimpleBenchPredicted

42.4 / 100

simpleqaPredicted

37.3

SWE-bench LitePredicted

33.8 / 100

tau-bench RetailPredicted

84.0 / 100

WebArenaPredicted

1.6 / 100

WildBenchPredicted

44.6

BFCL (Berkeley Function Calling)bfcl

55.5

Arc Agi 2epoch_ai

0.0

Epoch Capabilities Indexepoch_ai

145.0

FictionLiveBenchepoch_ai

0.6

OTIS Mock AIME 2024–2025epoch_ai

0.8

SimpleQA Verified (Epoch)epoch_ai

0.2

LiveBench Codinglivebench

68.2 / 100

LiveBench Data Analysislivebench

55.2 / 100

LiveBench Languagelivebench

75.5 / 100

LiveBench Mathlivebench

82.2 / 100

LiveBench Overalllivebench

65.9 / 100

LiveBench Reasoninglivebench

68.3 / 100

SWE-bench Verifiedswebench

56.2 / 100

Vectara Factual Consistencyvectara_hallucination

85.3 / 100

Vectara Hallucination Ratevectara_hallucination

14.7 / 100

WeirdMLweirdml

52.7 / 100