pytorch/rl

View on GitHub

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

Stars: 3.6k
Forks: 484
Open beginner issues: 0
Indexed issues: 187
Avg merge: 1d 1h
Merged PRs (30d): 207

Dominant language: Python
License: MIT
Last GitHub push: Sep 19, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: good first issue

Beginner friendly Everything

Open All

187 open issues indexed

[Feature Request] Add Support for FQL Open

enhancement

pytorch/rl#4433 · 1 reaction · 1 assignee · Opened Sep 19, 2026
[RFC] Introducing transform primatives for sim2real Open

enhancement

pytorch/rl#4422 · 4 comments · 1 assignee · Opened Sep 17, 2026
[BUG] SACTrainer and CQLTrainer unnecessarily query collector environments Open

pytorch/rl#4376 · 1 comment · 1 assignee · Opened Sep 14, 2026
[Feature Request] Track and shrink Hydra Config/class kwarg parity gaps Open

pytorch/rl#4228 · 1 comment · 1 assignee · Opened Sep 5, 2026
[BUG] GRPOLoss masking strategies produce incompatible shapes Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/rl#4227 · 3 comments · Opened Sep 5, 2026
[Feature Request] Transformers for first-class support in torchRL Open

enhancement

pytorch/rl#4192 · 1 comment · 1 assignee · Opened Sep 1, 2026
[RFC] Allow Trainer to work with distributed Learner implementations Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/rl#4185 · 1 comment · Opened Aug 29, 2026
Agents are bad at writing tests Open

bug

pytorch/rl#4144 · 5 comments · 1 assignee · Opened Aug 23, 2026
Intermittent segfault in libcuda.so.1 during PPO training with custom EnvBase + ParallelEnv workers Open

pytorch/rl#4142 · 7 comments · 1 assignee · Opened Aug 22, 2026
Bug: ValueError: cannot find context for 'fork' breaks 10 tutorials on Windows Open

bug

pytorch/rl#4122 · 1 comment · 1 assignee · Opened Aug 17, 2026
[Feature] WS1: Stable Component Boundaries for External Loops Open

enhancement

pytorch/rl#4092 · 2 comments · 2 assignees · Opened Aug 11, 2026
[Feature Request] AsyncEnvPool v2 Open

enhancement

pytorch/rl#4061 · 8 comments · 1 reaction · 2 assignees · Opened Aug 5, 2026
Prepare TorchRL's Hydra entry points for Hydra 1.4 Open

Difficulty 4/5 3-5 days Newbie friendliness 68/100

pytorch/rl#4055 · 2 comments · Opened Jul 31, 2026
[CI] Flaky Test Tracker workflow is failing Open

CI flaky-tracker-failure

pytorch/rl#4054 · 3 comments · 1 assignee · Opened Jul 30, 2026
[Feature Request] Add generation-stamped replay indices for safe delayed priority updates Open

enhancement

pytorch/rl#4041 · 18 comments · 1 reaction · 1 assignee · Opened Jul 22, 2026
[Feature Request] Add generation-safe conditional updates for mutable replay fields Open

enhancement

pytorch/rl#4040 · 10 comments · 1 reaction · 1 assignee · Opened Jul 22, 2026
[Feature Request] Separate replay selection distributions from sampling units and boundary policies Open

enhancement

pytorch/rl#4039 · 6 comments · 1 reaction · 1 assignee · Opened Jul 22, 2026
[RFC] Add a general monitoring and logger attachment API for collectors and replay buffers Open

Collectors Discussion needed enhancement Logging ReplayBuffers Services

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/rl#4012 · 6 comments · 2 reactions · Opened Jul 15, 2026
[Feature Request] Reward-hacking / training-instability monitoring transform Open

Difficulty 5/5 Over a week Newbie friendliness 38/100

pytorch/rl#3984 · 2 comments · Opened Jul 10, 2026
[RFC] Post-training efforts Open

enhancement

pytorch/rl#3948 · 4 comments · 2 assignees · Opened Jul 7, 2026
[RFC] Unify device configuration across collectors, evaluators, and the inference server Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 30/100

pytorch/rl#3943 · 1 comment · Opened Jul 6, 2026
[Feature] CI Checks for Stale PRs Open

enhancement

pytorch/rl#3906 · 1 comment · 1 assignee · Opened Jun 24, 2026
[Flaky Tests] 5686 flaky test(s) detected Open

CI flaky-test-tracker

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/rl#3865 · 1 comment · Opened Jun 13, 2026
[Feature Request] Mujoco Playground Integration (with multiagent support) Open

enhancement

pytorch/rl#3733 · 2 comments · 1 assignee · Opened May 9, 2026
[Feature Request] Add ACT Policy for Robotics Open

enhancement

pytorch/rl#3636 · 2 comments · 2 reactions · 1 assignee · Opened Apr 15, 2026
Feature request: public async evaluator API for dedicated eval devices Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/rl#3610 · Opened Apr 9, 2026
[Feature Request] Auto-batching inference server Open

enhancement

pytorch/rl#3462 · 6 comments · 1 assignee · Opened Feb 6, 2026
[Feature Request] Standardized API for Explicit Multi-Agent Communication Actions Open

enhancement

pytorch/rl#3391 · 1 comment · 1 assignee · Opened Jan 26, 2026
[BUG] `LineariseRewards` causes `KeyError` in `Compose` transform. Open

bug

pytorch/rl#3237 · 1 comment · 1 assignee · Opened Nov 20, 2025
[BUG] SliceSampler doesn't work as expected when collecting data from parallel environment? Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/rl#3194 · 8 comments · Opened Oct 14, 2025
[Feature Request] Question about LSTMModules Open

enhancement

pytorch/rl#3147 · 5 comments · 1 assignee · Opened Sep 4, 2025
[BUG] Struggling to reproduce GSM8K GRPO Training on B200 with CUDA 12.8. Open

bug

pytorch/rl#3141 · 1 assignee · Opened Aug 27, 2025
Announcement: PyTorch org (and rl) will be offered to PyTorch Foundation Open

Difficulty 5/5 Over a week Newbie friendliness 1/100

pytorch/rl#3130 · 3 reactions · Opened Aug 14, 2025
[BUG] AtariDQNExperienceReplay shows AccessDeniedException Open

bug

pytorch/rl#3113 · 1 comment · 1 assignee · Opened Aug 6, 2025
[Feature Request] Documentation explaining the recommended way to save and load agents Open

enhancement

pytorch/rl#3044 · 3 comments · 2 assignees · Opened Jul 8, 2025
[Feature Request] Documentation about weight update propagation Open

enhancement

pytorch/rl#3032 · 7 comments · 1 reaction · 2 assignees · Opened Jul 3, 2025
Comment/question about the integration of a PettingZoo AEC environment with TorchRL Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pytorch/rl#2990 · 1 comment · Opened Jun 6, 2025
Feature Request: Expand ChatEnv with Popular Agent Environment Integrations Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/rl#2984 · Opened Jun 2, 2025
[BUG] TensorSpecs require memory proportionally to batch size Open

bug

pytorch/rl#2974 · 1 comment · 1 assignee · Opened May 29, 2025
[Feature Request] MultiAgentLSTM Module and Full Example Open

enhancement

pytorch/rl#2911 · 4 comments · 1 assignee · Opened Apr 22, 2025
[Feature Request] Dynamically Update Environment Parameters (and Reset Transforms) in MultiaSyncDataCollector Workers Open

enhancement

pytorch/rl#2896 · 2 comments · 1 reaction · 1 assignee · Opened Apr 8, 2025
[BUG] RuntimeError when passing dialogue data to LLMEnv Open

bug

pytorch/rl#2875 · 1 comment · 1 assignee · Opened Mar 25, 2025
[Feature Request] A collector designed for LLMs Open

enhancement

pytorch/rl#2872 · 1 reaction · 1 assignee · Opened Mar 25, 2025
[BUG] DataCollectors fail when device is set to MPS Open

bug

pytorch/rl#2858 · 2 reactions · 1 assignee · Opened Mar 18, 2025
No example for UnityMLAgentsEnv or Wrapper for single or multiagent training Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/rl#2781 · 11 comments · Opened Feb 12, 2025
[Feature Request] Allow CUDNN based modules in losses Open

enhancement

pytorch/rl#2741 · 4 comments · 1 assignee · Opened Feb 3, 2025
[BUG] Maniskill3 crashes on D2H transfer after env rollout Open

bug

pytorch/rl#2739 · 1 comment · 1 assignee · Opened Feb 1, 2025
[Feature Request] Add Maniskill3 Custom Env Open

enhancement

pytorch/rl#2736 · 5 comments · 1 reaction · 1 assignee · Opened Jan 30, 2025
[BUG] SafeProbabilisticModule constructor missing `log_prob_keys` argument Open

bug

pytorch/rl#2731 · 1 assignee · Opened Jan 29, 2025
[Feature Request] Provide documentation on how to use CatFrames with a data collector and replay buffer for images Open

enhancement

pytorch/rl#2618 · 1 comment · 2 assignees · Opened Nov 29, 2024

pytorch/rl

Get new issues in your inbox