Science Domain Evals

45 resources · page 1 of 2

• ARC-AGI leaderboard

  Creator not supplied; see original source.

  ARC Prize leaderboard: model scores on ARC-AGI abstraction benchmarks plotted against cost per task.

  Not applicable (entity) · unknown · unknown resource · en

• BEIR

  Creator not supplied; see original source.

  BEIR GitHub repo: heterogeneous information-retrieval benchmark of many datasets for zero-shot evaluation of retrievers.

  Not applicable (entity) · unknown · unknown resource · en

• CAMEO

  Creator not supplied; see original source.

  CAMEO platform: weekly automated, blind benchmarking of protein structure prediction servers against newly released structures.

  Not applicable (entity) · unknown · unknown resource · en

• CASP Prediction Center

  Creator not supplied; see original source.

  CASP Prediction Center: blind community assessment of protein structure prediction methods.

  Not applicable (entity) · unknown · unknown resource · en

• CASP17

  Creator not supplied; see original source.

  CASP17 site: the 2026 round of the Critical Assessment of protein Structure Prediction, with targets and assessment results.

  Not applicable (entity) · unknown · unknown resource · en

• DREAM Challenges

  Creator not supplied; see original source.

  Sage Bionetworks page for DREAM Challenges: community biomedical challenges with held-out data and evaluation protocols.

  Not applicable (entity) · unknown · unknown resource · en

• HealthBench Professional: Benchmark review (Epoch AI)

  Creator not supplied; see original source.

  Epoch AI review of HealthBench Professional, auditing sampled tasks for errors and assessing the benchmark's reliability.

  Not applicable (entity) · unknown · unknown resource · en

• FinanceBench

  Creator not supplied; see original source.

  FinanceBench GitHub repo from Patronus AI: open-book financial question answering over public company filings.

  Not applicable (entity) · unknown · unknown resource · en

• FinanceBench (Patronus AI)

  Creator not supplied; see original source.

  Patronus AI announcement of FinanceBench, question answering over public financial filings to measure LLM failures.

  Not applicable (entity) · unknown · unknown resource · en

• FrontierMath

  Creator not supplied; see original source.

  Epoch AI FrontierMath: benchmark of unpublished expert-level mathematics problems in tiers, with model results.

  Not applicable (entity) · unknown · unknown resource · en

• GPQA

  Creator not supplied; see original source.

  GPQA GitHub repo: graduate-level, Google-proof science questions with baselines for evaluating expert reasoning.

  Not applicable (entity) · unknown · unknown resource · en

• GSM8K

  Creator not supplied; see original source.

  GSM8K GitHub repo from OpenAI: grade-school math word problems with solutions and evaluation code; archived.

  Not applicable (entity) · unknown · unknown resource · en

• GuacaMol

  Creator not supplied; see original source.

  BenevolentAI GuacaMol GitHub repo: benchmarks for generative chemistry models.

  Not applicable (entity) · unknown · unknown resource · en

• LAB-Bench

  Creator not supplied; see original source.

  LAB-Bench GitHub repo from FutureHouse: practical biology-research tasks such as literature, sequence and protocol questions.

  Not applicable (entity) · unknown · unknown resource · en

• LegalBench

  Creator not supplied; see original source.

  LegalBench site from Stanford: collaboratively built benchmark of legal-reasoning tasks for evaluating LLMs on US law.

  Not applicable (entity) · unknown · unknown resource · en

• LiveMedBench

  Creator not supplied; see original source.

  LiveMedBench paper page: a continuously refreshed medical LLM benchmark built from real clinical cases with automated rubric scoring.

  Not applicable (entity) · unknown · unknown resource · en

• Matbench

  Creator not supplied; see original source.

  Materials Project Matbench GitHub repo: benchmark tasks for materials property prediction.

  Not applicable (entity) · unknown · unknown resource · en

• Matbench Discovery

  Creator not supplied; see original source.

  Matbench Discovery GitHub repo: evaluation framework for ML models in materials stability discovery.

  Not applicable (entity) · unknown · unknown resource · en

• MATH

  Creator not supplied; see original source.

  MATH dataset GitHub repo: competition mathematics problems with step-by-step solutions for evaluating reasoning.

  Not applicable (entity) · unknown · unknown resource · en

• MathArena

  Creator not supplied; see original source.

  MathArena: evaluation of LLMs on newly released math competitions and arXiv-derived problems to avoid contamination.

  Not applicable (entity) · unknown · unknown resource · en

• MedBench

  Creator not supplied; see original source.

  OpenCompass MedBench portal: Chinese medical benchmark platform for evaluating LLMs on clinical knowledge and tasks.

  Not applicable (entity) · unknown · unknown resource · zh

• MedHELM

  Creator not supplied; see original source.

  Stanford CRFM MedHELM: evaluation of language models on clinician-defined medical tasks beyond exam-style questions.

  Not applicable (entity) · unknown · unknown resource · en

• MedHELM (Stanford CRFM)

  Creator not supplied; see original source.

  Latest MedHELM results page from Stanford CRFM, scoring language models on a taxonomy of real clinical tasks.

  Not applicable (entity) · unknown · unknown resource · en

• MedMNIST

  Creator not supplied; see original source.

  MedMNIST site: standardised 2D and 3D biomedical image classification datasets for lightweight benchmarking.

  Not applicable (entity) · unknown · unknown resource · en

• MERA Industrial

  Creator not supplied; see original source.

  MERA Industrial: benchmark of foundation models on industry-specific Russian-language tasks.

  Not applicable (entity) · unknown · unknown resource · ru

• MOSES

  Creator not supplied; see original source.

  MOSES GitHub repo: benchmarking platform with metrics and baselines for molecular generation models.

  Not applicable (entity) · unknown · unknown resource · en

• Open Catalyst Project

  Creator not supplied; see original source.

  Open Catalyst Project site: atomistic datasets, models and challenges for AI-based catalyst discovery.

  Not applicable (entity) · unknown · unknown resource · en

• Open Graph Benchmark

  Creator not supplied; see original source.

  Open Graph Benchmark site from Stanford: graph datasets, data loaders and evaluators for graph machine learning.

  Not applicable (entity) · unknown · unknown resource · en

• frontier-evals

  Creator not supplied; see original source.

  OpenAI frontier-evals GitHub repo: code for evaluations such as PaperBench and SWE-Lancer.

  Not applicable (entity) · unknown · unknown resource · en

• PhysioNet Challenges

  Creator not supplied; see original source.

  PhysioNet Challenges page: annual open challenges on clinical physiological-signal analysis with scoring methods and data.

  Not applicable (entity) · unknown · unknown resource · en

• ProteinGym

  Creator not supplied; see original source.

  ProteinGym: benchmark and leaderboard for predicting the effects of protein mutations.

  Not applicable (entity) · unknown · unknown resource · en

• Robot learning leaderboard (RoboArena)

  Creator not supplied; see original source.

  Robot-learning.ml leaderboard site for evaluating robot learning policies.

  Not applicable (entity) · unknown · unknown resource · und

Plain-text Markdown copy