PicadoLabs/agent-bench
A local-first, provider-agnostic platform to benchmark, score, and diagnose autonomous AI coding agents on real-world polyglot software tasks.
GitHub repository with 5 stars and 3 forks.
Language: Python
Topics: agentic-ai, ai-agents, autonomous-agents, benchmarking, cli, code-generation, developer-tools, llm-evals, local-first, python