← Back to directory

claw-bench

Measure AI agents’ performance with standardized tests across 314 tasks, 33 domains, and 4 difficulty levels for clear, reproducible comparison.

agentagentic-evaluationaiai-scientistai4scienceauto-researchbenchmarkbenchmarksclaudeclawdbot

Setup / get started

Start from the GitHub repository: source code and README for claw-bench.

Alternatives to claw-bench

Similar projects in this directory:

  • ResearchClawBench — ResearchClawBench: Evaluating AI Agents for Automated Research from Re-Discovery to New-Discovery
  • smartness-eval — 🎯 14-Dimension AI Agent Intelligence Assessment | 14维度 AI Agent 智能度自动评估技能 | OpenClaw Skill | Aligned with…
  • NanoResearch — 🦞+🔬: NanoResearch: The Autonomous AI Research Assistant
  • clawdbot-skill-parallel — High-accuracy web research via Parallel.ai API - outperforms Perplexity and Exa on benchmarks. Rich excerpts…
  • nopua — Provide tools to identify and counteract pickup artist tactics using data-driven benchmarks and clear evidence for…
By wild-balthazar224Added 3/30/2026