AI Intelligence · Benchmarks

AI Benchmark Center

Live leaderboards across reasoning, coding, math, expert knowledge, and agent performance.

Current Leaders by Category

💻

Coding

CoreDev X

97.5

🛠

SWE-bench

CoreDev X

63.8

🧠

Reasoning

Athena Logic Pro

96.5

🔬

Science

Athena Logic Pro

92.1

Overall Model Ranking

(avg. normalized score across all benchmarks)

🥇NEW

CoreDev X

Axiom Dynamics

100.0

avg. score

🥈NEW

Athena Logic Pro

WisdomWorks

100.0

avg. score

🥉NEW

Aurora Prime v2

NovaSense AI

99.7

avg. score

4NEW

Apex Intellect

Veridian Labs

99.5

avg. score

5NEW

Sentient Core X

CogniPath

99.2

avg. score

6NEW

OmniVision AGI

Synthetica Inc.

98.9

avg. score

7NEW

TerraNova AGI

GlobalMind Inc.

98.8

avg. score

8NEW

ProtoGenius Code v1.0

SyntheDev

98.8

avg. score

💻

Coding

HumanEval · % pass@1

164 programming challenges requiring correct code from docstrings.

🥇

CoreDev XAxiom DynamicsNEW

Aug 5, 2026

97.5

🥈

ProtoGenius Code v1.0SyntheDevNEW

Aug 1, 2026

97.2

🥉

Genesis Code v3Origin LabsNEW

Aug 2, 2026

OmniCraft Coder v2.1AetherMindNEW

Jul 29, 2026

96.8

Arcana EngineerMysticTechNEW

Aug 4, 2026

96.3

🛠

SWE-bench

SWE-bench · % resolved

Real GitHub issues from popular open-source repositories.

🥇

CoreDev XAxiom DynamicsNEW

Aug 5, 2026

63.8

🥈

ProtoGenius Code v1.0SyntheDevNEW

Aug 1, 2026

62.5

🥉

Genesis Code v3Origin LabsNEW

Aug 2, 2026

61.5

OmniCraft Coder v2.1AetherMindNEW

Jul 29, 2026

60.1

Arcana EngineerMysticTechNEW

Aug 4, 2026

59.5

🧠

Reasoning

MMLU · % accuracy

57-subject knowledge test across STEM, humanities, and social sciences.

🥇

Athena Logic ProWisdomWorksNEW

Aug 7, 2026

96.5

🥈

Aurora Prime v2NovaSense AINEW

Aug 18, 2026

96.3

🥉

Apex IntellectVeridian LabsNEW

Aug 1, 2026

96.1

Sentient Core XCogniPathNEW

Aug 25, 2026

Sentinel AGI 1.0LuminaTechNEW

Aug 5, 2026

95.8

🔬

Science

GPQA · % accuracy

PhD-level science questions in biology, chemistry, and physics.

🥇

Athena Logic ProWisdomWorksNEW

Aug 7, 2026

92.1

🥈

Aurora Prime v2NovaSense AINEW

Aug 18, 2026

91.8

🥉

Apex IntellectVeridian LabsNEW

Aug 1, 2026

91.5

Sentient Core XCogniPathNEW

Aug 25, 2026

91.1

TerraNova AGIGlobalMind Inc.NEW

Aug 9, 2026

90.9

About These Benchmarks

MMLU (Massive Multitask Language Understanding)

Tests knowledge across 57 subjects including STEM, humanities, and social sciences. 14,000+ questions.

HumanEval (Coding)

164 hand-crafted programming challenges. Measures ability to produce correct code from docstrings.

MATH

12,500 competition math problems from AMC, AIME, and AMC 10/12. Tests advanced mathematical reasoning.

SWE-bench Verified

Real GitHub issues from popular open-source repos. Measures end-to-end software engineering capability.