Seed-OSS-36B-Instruct

August 20, 2025Last synced Apr 7, 2026, 4:00 PM

Blended Price

$0.30/M

Input Price

$0.21/M

Output Price

$0.57/M

Speed

44 tok/s

TTFT

1.71s

Benchmark Scores

Intelligence Index

External Benchmarks

AA-Omniscience Accuracy

How Seed-OSS-36B-Instruct 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

0.8 / 100

Terminal-Bench Hard

0.1 / 100

τ²-Bench

0.5 / 100

AA-Omniscience Hallucination RatePredicted

96.3 / 100

Aider PolyglotPredicted

63.7 / 100

AIMEPredicted

0.6 / 30

AlpacaEval 2.0 LCPredicted

32.9 / 100

AlpacaEval 2.0 RawPredicted

25.6 / 100

ARC-AGI-1Predicted

26.2 / 100

ARC-AGI-1 Cost per TaskPredicted

0.1

ARC-AGI-2Predicted

0.5 / 100

ARC-AGI-2 Cost per TaskPredicted

0.2

BFCL (Berkeley Function Calling)Predicted

49.6

BigCodeBench CompletePredicted

54.2 / 100

BigCodeBench InstructPredicted

44.2 / 100

AA Intelligence Index (Matrix)Predicted

66.7

AA Long Context Reasoning (Matrix)Predicted

72.2

AIME 2024Predicted

94.2

AIME 2025 (Matrix)Predicted

95.4

Arena-Hard AutoPredicted

80.2

BrowseCompPredicted

62.9

BRUMO 2025Predicted

92.8

CMIMC 2025Predicted

80.4

GPQA Diamond (Matrix)Predicted

81.5

HLE (Matrix)Predicted

20.7

HMMT Feb 2025Predicted

77.4

HMMT Nov 2025Predicted

90.3

HumanEvalPredicted

90.9

IFBench (Matrix)Predicted

50.1

IFEvalPredicted

89.0

IMO 2025Predicted

20.3

LiveCodeBench (Matrix)Predicted

73.6

MATH-500 (Matrix)Predicted

97.3

MathArena Apex 2025Predicted

1.3

MMLU-Pro (Matrix)Predicted

81.5

MMMU-ProPredicted

81.4

MRCR v2Predicted

76.8

OSWorldPredicted

42.8

SimpleQAPredicted

35.4

SMT 2025Predicted

84.9

SWE-bench ProPredicted

39.3

Tau-Bench Telecom (Matrix)Predicted

96.6

Terminal-Bench 2.0Predicted

27.5

Terminal-Bench 1.0Predicted

24.6

USAMO 2025Predicted

13.4

Video-MMUPredicted

86.5

browsecompPredicted

60.0

BullshitBenchPredicted

21.6 / 100

Aider PolyglotPredicted

0.0

Apex AgentsPredicted

2.3

Arc Agi 2Predicted

0.0

BIG-Bench HardPredicted

3.0

CAD-EvalPredicted

3.7

Chess PuzzlesPredicted

0.2

CyBenchPredicted

0.3

DeepResearchBenchPredicted

0.4

FictionLiveBenchPredicted

0.6

GeoBenchPredicted

0.0

GSM8K (Epoch)Predicted

4.6

HellaSwagPredicted

0.3

Lech Mazur WritingPredicted

7.9

METR Time HorizonsPredicted

2.1

OTIS Mock AIME 2024–2025Predicted

0.5

PosttrainbenchPredicted

0.0

SimpleQA Verified (Epoch)Predicted

0.3

The Agent CompanyPredicted

2.2

TriviaQAPredicted

6.5

WinoGrandePredicted

0.9

FrontierMathPredicted

16.2 / 100

GAIA Level 1Predicted

70.6

GAIA Level 2Predicted

55.0

GAIA Level 3Predicted

38.1

GAIAPredicted

54.1 / 100

LegalBenchPredicted

87.0 / 100

LiveBench CodingPredicted

68.4 / 100

LiveBench Data AnalysisPredicted

48.7 / 100

LiveBench LanguagePredicted

59.4 / 100

LiveBench MathPredicted

67.4 / 100

LiveBench OverallPredicted

52.4 / 100

LiveBench ReasoningPredicted

49.7 / 100

LongBench v2 EasyPredicted

52.9

LongBench v2 HardPredicted

46.5

LongBench v2Predicted

42.1 / 100

MATH-500Predicted

1.0 / 100

MathVistaPredicted

62.3 / 100

MedQA (USMLE)Predicted

92.1

MLE-benchPredicted

30.9 / 100

MMMUPredicted

77.8 / 100

MMTU Table UnderstandingPredicted

60.4 / 100

MT-BenchPredicted

7.9 / 10

NoLiMa (NIAH)Predicted

94.1 / 100

OCRBench v2Predicted

84.2 / 100

Open LLM AveragePredicted

27.0 / 100

Open LLM: BBHPredicted

56.7 / 100

Open LLM: GPQAPredicted

32.0 / 100

Open LLM: IFEvalPredicted

51.6 / 100

Open LLM: MATH Level 5Predicted

18.5 / 100

Open LLM: MMLU-PROPredicted

40.7 / 100

Open LLM: MUSRPredicted

42.7 / 100

RE-BenchPredicted

32.7

SimpleBenchPredicted

39.6 / 100

simpleqaPredicted

34.8

SWE-bench LitePredicted

37.6 / 100

SWE-bench VerifiedPredicted

60.7 / 100

tau-bench RetailPredicted

80.9 / 100

Vectara Factual ConsistencyPredicted

91.7 / 100

Vectara Hallucination RatePredicted

8.3 / 100

WebArenaPredicted

3.4 / 100

WeirdMLPredicted

46.4 / 100

WildBenchPredicted

48.8

Chatbot Arena ELOchatbot_arena

669.0