Models Agents EvalsIndustry

AgMoDB by @mistakeknot

Llama 3.2 Instruct 3B | AgMoDB

Llama 3.2 Instruct 3B

September 25, 2024Last synced Apr 7, 2026, 4:00 PM

Blended Price

$0.085/M

Input Price

$0.085/M

Output Price

$0.085/M

Speed

55 tok/s

TTFT

0.33s

Benchmark Scores

Intelligence Index

External Benchmarks

AA-Omniscience Accuracy

How Llama 3.2 Instruct 3B 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

9.7

Coding Index

—

Math Index

3.3

MMLU Pro

0.3 / 100

GPQA Diamond

0.3 / 100

HLE

0.1 / 100

LiveCodeBench

0.1 / 100

SciCode

0.1 / 100

MATH-500

0.5 / 100

AIME

0.1 / 30

AIME 2025

0.0 / 30

IFBench

0.3 / 100

LCR

0.0 / 100

Terminal-Bench Hard

— / 100

τ²-Bench

0.2 / 100

Predicted

48.8 / 100

AA-Omniscience Hallucination RatePredicted

97.0 / 100

Aider PolyglotPredicted

30.8 / 100

AlpacaEval 2.0 LCPredicted

19.5 / 100

AlpacaEval 2.0 RawPredicted

15.6 / 100

ARC-AGI-1Predicted

60.5 / 100

ARC-AGI-1 Cost per TaskPredicted

0.2

ARC-AGI-2Predicted

7.5 / 100

ARC-AGI-2 Cost per TaskPredicted

0.3

BFCL (Berkeley Function Calling)Predicted

29.2

BigCodeBench CompletePredicted

30.8 / 100

BigCodeBench InstructPredicted

24.1 / 100

AA Intelligence Index (Matrix)Predicted

53.2

AA Long Context Reasoning (Matrix)Predicted

76.3

AIME 2024Predicted

39.8

AIME 2025 (Matrix)Predicted

92.5

Arena-Hard AutoPredicted

28.3

BrowseCompPredicted

84.3

BRUMO 2025Predicted

99.5

CMIMC 2025Predicted

91.8

CritPtPredicted

8.5

GPQA Diamond (Matrix)Predicted

59.0

GSM8KPredicted

90.3

HLE (Matrix)Predicted

23.0

HMMT Feb 2025Predicted

77.6

HMMT Nov 2025Predicted

93.8

HumanEvalPredicted

66.4

IFBench (Matrix)Predicted

42.5

IFEvalPredicted

80.1

IMO 2025Predicted

41.5

LiveCodeBench (Matrix)Predicted

46.1

MATH-500 (Matrix)Predicted

90.9

MathArena Apex 2025Predicted

12.1

MMLUPredicted

66.4

MMLU-Pro (Matrix)Predicted

47.2

MMMU-ProPredicted

77.7

MRCR v2Predicted

81.0

OSWorldPredicted

44.2

SimpleQAPredicted

38.1

SMT 2025Predicted

85.2

SWE-bench ProPredicted

33.4

Tau-Bench Telecom (Matrix)Predicted

98.8

Terminal-Bench 2.0Predicted

68.7

Terminal-Bench 1.0Predicted

32.0

USAMO 2025Predicted

13.0

Video-MMUPredicted

84.5

browsecompPredicted

85.8

BullshitBenchPredicted

23.7 / 100

Aider PolyglotPredicted

0.2

Apex AgentsPredicted

4.0

Arc Agi 2Predicted

0.1

BALROGPredicted

0.0

BIG-Bench HardPredicted

3.0

BoolQPredicted

0.8

CAD-EvalPredicted

6.3

Chess PuzzlesPredicted

0.3

CyBenchPredicted

0.2

DeepResearchBenchPredicted

0.4

FictionLiveBenchPredicted

0.6

GdpvalPredicted

0.5

GeoBenchPredicted

0.0

GSM8K (Epoch)Predicted

4.8

GSOPredicted

0.4

HellaSwagPredicted

0.0

HlePredicted

0.2

Lech Mazur WritingPredicted

7.3

METR Time HorizonsPredicted

9.5

OTIS Mock AIME 2024–2025Predicted

0.1

PIQAPredicted

0.9

PosttrainbenchPredicted

0.0

SimpleQA Verified (Epoch)Predicted

0.5

The Agent CompanyPredicted

1.1

TriviaQAPredicted

15.8

VPCTPredicted

0.5

WinoGrandePredicted

0.7

FrontierMathPredicted

30.4 / 100

GAIA Level 1Predicted

15.8

GAIA Level 2Predicted

2.0

GAIA Level 3Predicted

0.0

GAIAPredicted

9.3 / 100

LegalBenchPredicted

35.6 / 100

LiveBench CodingPredicted

76.8 / 100

LiveBench Data AnalysisPredicted

66.7 / 100

LiveBench LanguagePredicted

78.3 / 100

LiveBench MathPredicted

84.4 / 100

LiveBench OverallPredicted

71.1 / 100

LiveBench ReasoningPredicted

74.9 / 100

LongBench v2 EasyPredicted

34.4

LongBench v2 HardPredicted

28.6

LongBench v2Predicted

32.1 / 100

MathVistaPredicted

47.3 / 100

MedQA (USMLE)Predicted

57.8

MLE-benchPredicted

58.2 / 100

MMMUPredicted

62.2 / 100

MMTU Table UnderstandingPredicted

56.8 / 100

MT-BenchPredicted

7.7 / 10

NoLiMa (NIAH)Predicted

85.4 / 100

OCRBench v2Predicted

62.5 / 100

RE-BenchPredicted

99.9

SimpleBenchPredicted

38.1 / 100

simpleqaPredicted

35.3

SWE-bench LitePredicted

14.5 / 100

SWE-bench VerifiedPredicted

48.5 / 100

tau-bench RetailPredicted

90.8 / 100

Terminal-Bench HardPredicted

0.0 / 100

Vectara Factual ConsistencyPredicted

87.8 / 100

Vectara Hallucination RatePredicted

12.2 / 100

WebArenaPredicted

0.0 / 100

WeirdMLPredicted

27.3 / 100

WildBenchPredicted

30.9

Open LLM Averageopen_llm_leaderboard

17.3 / 100

Open LLM: BBHopen_llm_leaderboard

43.3 / 100

Open LLM: GPQAopen_llm_leaderboard

26.9 / 100

Open LLM: IFEvalopen_llm_leaderboard

47.3 / 100

Open LLM: MATH Level 5open_llm_leaderboard

9.8 / 100

Open LLM: MMLU-PROopen_llm_leaderboard

29.2 / 100

Open LLM: MUSRopen_llm_leaderboard

35.0 / 100