// Evaluate & guard
OSWorldv0.1.16
[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
Open source Maintained For developers
IN ONE PARAGRAPH
OSWorld is [NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments.
Good for: Developers and teams who want to evaluate & guard.
01How to start
- 1Read the official quick-start (link above).
- 2Install or sign up as the official guide describes.
- 3Try it on a small example before wiring it into a project.
02Alternatives & comparison
Compare side by side → lmms-eval One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio Tasks ★ 4.5k opencompass OpenCompass is an LLM evaluation platform, supporting a wide range of models from OpenAI, Anthropic, Gemini, Qwen, GLM, DeepSeek, etc, across 100+ datasets covering knowledge, reasoning, coding, science, language, long-context, and safety. ★ 7.5k SWE-bench SWE-bench: Can Language Models Resolve Real-world Github Issues? ★ 6.1k tau2-bench τ-Bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains ★ 2.2k promptfoo Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, DeepSeek, and more. Simple declarative configs with command line and CI/CD integration. Used by OpenAI and Anthropic. ★ 26k
03Related
Versions, why it is listed, for AI to read
Recent releases
For AI agents
- Interfaces：CLI
- Machine-readable record：/api/v1/software/osworld.json · llms.txt
- Added 2026-10-10 · updated 2026-10-22