OpenNSWM-Lab/FAROS
Blueprint-driven AutoResearch runtime that orchestrates an LLM research workflow from idea refinement through experiment scaffolding, LaTeX paper drafting and simulated peer review.
Awesome AI Agents › Research Platforms & Simulators
aipoch/open-science
AIPOCH Open-Science is an AI research workbench aimed at making computational science work reproducible, distributed as an open-source, local-first and model-agnostic desktop application for macOS, Windows and Linux. Being local-first and model-agnostic means a researcher's data and workflow stay on their own machine and are not tied to a single model provider, in contrast to hosted research-assistant products. The project publishes a DOI through Zenodo for citing specific releases, indicating it is intended to be referenced in academic work the way a piece of research software normally would be. It highlights a top ranking on the public 50-sample split of BiomniBench-DA, a benchmark for data-analysis agents, as evidence of its capability on realistic scientific data-analysis tasks rather than only synthetic examples. Documentation is available in English plus Simplified and Traditional Chinese, and the project maintains a Discord community and a dedicated marketing site in addition to its GitHub releases. As a category, it positions itself between a general AI coding agent and a domain-specific scientific tool: the workbench framing suggests it is built around the recurring steps of a research workflow (data analysis, experiment tracking, reproducibility) rather than being a narrow single-purpose script, but the README excerpt available does not enumerate every specific workflow stage it automates beyond emphasizing local-first operation and benchmark performance on data-analysis tasks. It is licensed under Apache 2.0 and distributed as downloadable releases rather than a library or CLI package.
https://github.com/aipoch/open-science
Also in Research Platforms & Simulators
WooooDyy/AgentGym-RL
Framework for training LLM agents on long-horizon, multi-turn decision making with reinforcement learning across a broad set of interactive environments, plus the ScalingInter-RL training method.
Purewhiter/mobilegym
MobileGym is a browser-hosted, fully programmable Android-style simulation platform with 28 simulated apps and 416 task templates for benchmarking and reinforcement-learning training of mobile GUI agents.
SimWorld-AI/SimWorld
SimWorld is an Unreal Engine 5 based open-ended simulator for developing and evaluating LLM and VLM agents in realistic physical and social environments such as procedurally generated cities.
Tencent/AICGSecEval
A.S.E is Tencent's repository-level benchmark that measures the security of AI-generated code, building tasks from real GitHub projects and CVE patches and scoring output with static and dynamic analysis.
facebookresearch/meta-agents-research-environments
Meta's research platform for evaluating AI agents in dynamic simulated environments, hosting the Gaia2 benchmark of 800 evolving scenarios across ten universes, with a CLI, a web GUI and a public leaderboard.
HUST-AI-HYZ/MemoryAgentBench
Benchmark and evaluation harness that measures memory in LLM agents across accurate retrieval, test-time learning, long-range understanding and conflict resolution using incremental multi-turn interactions.
DigiRL-agent/digirl
DigiRL is a research project providing code and resources for training autonomous reinforcement learning agents to control Android devices in real-world environments using novel training algorithms and multiple training modes.