Benchmarks agents

PinchBench is a benchmarking system for evaluating LLM models as OpenClaw coding agents. Made with 🦀 by the humans at https://kilo.ai

Open-source benchmark for browser AI agents on daily tasks.

Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.

🦞 ResearchClawBench: Evaluating AI Agents for Automated Research from Re-Discovery to New-Discovery

Track 14,810+ AI benchmark, eval, dataset, and data-quality records from 37 public sources, with linked evidence and daily updates.

Comprehensive suite for evaluating perturbation prediction models

Benchmark for LLM-based Agents in Computational Biology

🦞 ClawMark: A Living-World Benchmark for Multi-Day, Multimodal Coworker Agents

A Comprehensive Rare Disease Diagnostic Dataset with nearly 50,000 patients covering more than 4000 diseases

Benchmark for evaluating LLM agents in bioinformatics

Opensource repository containing task and image data for DrugDiscoveryBench

Digital Longevity Research

Cell paper — Arc Institute community benchmark of zero-shot perturbation-response generalization across cellular contexts.

Benchmark for data-driven biomedical hypothesis validation — 1,029 hypothesis-centric tasks with 1,177 analysis plans curated from 300+ published studies on cBioPortal patient data. Includes non-verifiable hypotheses the data can neither support nor refute. Dataset on Hugging Face (ODbL); ships with the BioDSA framework.