StonyBrookNLP/appworld
AppWorld is a controllable environment for benchmarking function-calling and interactive coding AI agents.
Not currently ranked — collecting fresh signals.
star history
AppWorld provides a simulated world of applications and people to evaluate how well LLM-based agents can use tools, call functions, and write code interactively. It serves as a standardized benchmark for measuring the performance of coding agents and function-calling systems. The platform includes a task explorer, API explorer, and leaderboard for comparing different agent implementations on realistic software interaction scenarios.
Frequently asked
- What is StonyBrookNLP/appworld?
- AppWorld is a controllable environment for benchmarking function-calling and interactive coding AI agents.
- Is appworld open source?
- Yes — StonyBrookNLP/appworld is open source, released under the Apache-2.0 license.
- What language is appworld written in?
- StonyBrookNLP/appworld is primarily written in Python.
- How popular is appworld?
- StonyBrookNLP/appworld has 447 stars on GitHub.
- Where can I find appworld?
- StonyBrookNLP/appworld is on GitHub at https://github.com/StonyBrookNLP/appworld.