Claude 3.5 Sonnet vs DeepSeek V3 vs GPT-4o (2024-11-20)

Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.

FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.

Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.

Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.

MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.

Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.

OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.

SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.

WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.

Balrog · benchmarks AI agents on text-based adventure games, testing language understanding, strategic planning, and long-horizon reasoning.

CadEval · evaluates the ability to generate and reason about Computer-Aided Design code, testing spatial reasoning and engineering knowledge.

Cybench · evaluates AI on real Capture-The-Flag cybersecurity challenges, testing vulnerability analysis, exploitation, and security reasoning.

GeoBench · tests geographic knowledge and spatial reasoning across countries, landmarks, coordinates, and geopolitical understanding.

GSO-Bench · evaluates AI models on real-world open-source software engineering tasks, testing the ability to understand and resolve actual GitHub issues.

HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation.

The Agent Company · tests AI agents on realistic corporate tasks like email management, code review, data analysis, and cross-tool workflows.

VideoMME · multimodal benchmark testing video understanding across diverse domains, requiring temporal reasoning and cross-frame comprehension.