Meta Agents Research Environments (ARE)
MIT research platform from Meta for simulating dynamic, multi-step agent tasks, including official Gaia2 evaluation.
Tags
Description
Established Meta project (repo facebookresearch/meta-agents-research-environments; not archived; created 2025-09-11; last pushed 2026-09-30; not presented as a brand-new 2026 launch). Root LICENSE is MIT (Copyright Meta Platforms, Inc. and affiliates). Vendor: Meta. PyPI meta-agents-research-environments 1.2.0 verified 2026-11-13 (not yanked; files uploaded 2025-11-11; project URLs point at this repository). There were no GitHub Releases on 2026-11-13; because the repo has been pushed since that PyPI upload, this entry notes that the 1.2.0 wheels may lag main. README install is pip install meta-agents-research-environments (optional [gui] extra). Docs at https://facebookresearch.github.io/meta-agents-research-environments/ and the Gaia2 evaluation guide (both HTTP 200). README describes evolving simulated environments, multi-step tasks, and official Gaia2 evaluation (800 scenarios). A container-based Gaia2-CLI stack lives under gaia2-cli/. Simulated apps can include write actions (mail, contacts, and similar in-sim tools). Model-API and compute costs for evaluations are separate; no keys or benchmark runs were used.
Agent access
Related
- Gaia2: CC-BY-4.0 follow-up to GAIA: 800 dynamic, time-aware agent scenarios released on Hugging Face and run with Meta’s ARE framework.
- GAIA: Benchmark for general AI assistants ("Benchmarking General AI Agents"), published as a Hugging Face dataset under the gaia-benchmark org.
Sources
- https://github.com/facebookresearch/meta-agents-research-environments
- https://github.com/facebookresearch/meta-agents-research-environments/blob/main/LICENSE
- https://github.com/facebookresearch/meta-agents-research-environments/blob/main/README.md
- https://pypi.org/project/meta-agents-research-environments/
- https://facebookresearch.github.io/meta-agents-research-environments/
- https://facebookresearch.github.io/meta-agents-research-environments/user_guide/gaia2_evaluation.html
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)