Framework for evaluating and improving agents
-
Updated
Sep 1, 2026 - Python
#
rl-environments
Here are 30 public repositories matching this topic...
A curated, non-BS library of the best resources for building and evaluating AI agents — papers, blogs, talks, tools, benchmarks. Maintained by BenchFlow.
-
Updated
Aug 20, 2026
HuggingEnvs — RL Environments 101: building and scaling RL environments in the age of LLMs
-
Updated
Aug 30, 2026 - HTML
A Universal Platform for Training and Evaluation of Mobile Interaction
-
Updated
Sep 24, 2025 - Python
A graphical interface for reinforcement learning and gym-based environments.
-
Updated
Mar 24, 2023 - Python
Interoperating between (Deep) Reiforcement Learning libraries
reinforcement-learning deep-learning reinforcement-learning-algorithms rl flax jax reinforcement-learning-environments rl-environments
-
Updated
Mar 6, 2024 - Python
Gymnasium-style API standard for RL environment creation in JAX
-
Updated
May 19, 2026 - Python
Workspace manager for coding agents. Interactively solve and develop Harbor tasks.
-
Updated
Jul 8, 2026 - Python
Create new gridworld gym environments easily
-
Updated
Jul 2, 2020 - Python
Awesome Environment Scaling for AI Agents — a periodically updated survey and curated list of papers, projects, RL environments, world models, agent sandboxes, and open research problems
awesome reinforcement-learning survey awesome-list ai-agents world-models rl-environments agent-sandbox llm llm-agents agentic-ai agent-memory agent-environments synthetic-environments environment-scaling
-
Updated
Aug 31, 2026
Adversarial QA for LLM-RL environments: find out what reward an empty answer earns. Model-free, zero API cost.
python testing reinforcement-learning ci evaluation fuzzing pytest-plugin quality-assurance grader ai-safety red-teaming rl-environments llm rlhf llm-evaluation verifiers reward-hacking adversarial-testing prime-intellect openenv
-
Updated
Aug 14, 2026 - Python
Turn any real software into a replayable RL environment for training AI agents — deterministic replay, verifiable rewards, TRL & verifiers adapters.
python machine-learning reinforcement-learning ai-agents post-training rl-environments trl llm grpo verifiable-rewards agent-training
-
Updated
Aug 21, 2026 - Python
A configurable 3D jump-chess framework with A/B geometries, multi-player rule switches, and RL/self-play training support.
benchmarking benchmark simulation pathfinding multi-agent benchmarks simulations board-games spatial-reasoning ai-research game-environment rl-environment reinforcement-learning-environments rl-environments 3d-checkers-game multi-agent-rl chinese-checkers-ai
-
Updated
Aug 25, 2026 - Python
Agent-evaluation environments: planted-truth worlds, ungameable graders, calibrated difficulty
-
Updated
Jul 27, 2026 - Python
A lightweight, open-source framework that turns historical GitHub pull requests into reproducible, verifiable software-engineering tasks for training and evaluating coding agents.
python docker benchmark reinforcement-learning evaluation developer-tools software-engineering agents ai-agents rl-environments github-prs llm coding-agents swe-bench
-
Updated
Jun 13, 2026 - Python
Foundry Lite: a public runnable sample of Veyl’s local environment harness for software-engineering agent evals.
-
Updated
Jun 15, 2026 - Python
Open-source RL environments and evals for the capabilities we want AI to have. Judge-free scoring, mandatory baselines, and an enforced defensive-asymmetry gate.
-
Updated
Aug 10, 2026 - Python
Sound error bounds, symbolic GPU safety checks and targeted falsification for Triton kernels. 88% of planted bugs pass the standard fixed-shape allclose test; Litmus catches 100% with 0% false positives, on CPU.
verification property-based-testing triton formal-methods numerical-analysis gpu-kernels rl-environments reward-model
-
Updated
Aug 17, 2026 - Python
An original Terminal-Bench 3 task that Claude Opus 5 and GPT-5.6 both fail three times out of three, and the research on why. The identity model lives in the data rather than the spec, and the graded core is opaque, so an agent cannot check its own answer.
benchmark entity-resolution verification ai-agents rl-environments llm-evaluation agent-evaluation terminal-bench
-
Updated
Aug 27, 2026 - Python
Add this topic to your repo
To associate your repository with the rl-environments topic, visit your repo's landing page and select "manage topics."