-
Notifications
You must be signed in to change notification settings - Fork 480
Pull requests: pytorch/rl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[BugFix] Stamp AsyncBatchedCollector requests with an explicit exploration type
BugFix
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Collectors
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Modules
sota-implementations/
#4310
opened Sep 8, 2026 by
vmoens
Collaborator
Loading…
[Performance] DreamerV3: freeze the setup heap before the training loop
benchmarks/trigger
Run the full benchmark suite on main right after this PR merges
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Objectives
Performance
Performance issue or suggestion for improvement
sota-implementations/
#4308
opened Sep 8, 2026 by
vmoens
Collaborator
•
4/4
Loading…
[Performance] Keep the DreamerV3 replay write-back off the learner stream
benchmarks/trigger
Run the full benchmark suite on main right after this PR merges
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Objectives
Performance
Performance issue or suggestion for improvement
ReplayBuffers
sota-implementations/
#4307
opened Sep 8, 2026 by
vmoens
Collaborator
•
3/4
Loading…
[Performance] Serve process-slot inference passes from pinned staging batches
benchmarks/trigger
Run the full benchmark suite on main right after this PR merges
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Modules
Performance
Performance issue or suggestion for improvement
#4306
opened Sep 8, 2026 by
vmoens
Collaborator
•
2/4
Loading…
[Performance] Chunk process-slot worker results in AsyncBatchedCollector
benchmarks/trigger
Run the full benchmark suite on main right after this PR merges
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Collectors
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Performance
Performance issue or suggestion for improvement
#4305
opened Sep 8, 2026 by
vmoens
Collaborator
•
1/4
Loading…
[Feature] Serve the DreamerV3 acting policy from environment processes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Feature
New feature
Objectives
sota-implementations/
#4304
opened Sep 8, 2026 by
vmoens
Collaborator
Loading…
6 of 9 tasks
[Feature] Add LBForaging environment wrapper
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Environments
Adds or modifies an environment wrapper
Feature
New feature
#4301
opened Sep 8, 2026 by
Iliamsou
Loading…
[Performance] Compile the DreamerV3 learner losses and make the two-hot decode reduction-safe
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Integrations/torch_geometric
Integrations
Modules
Objectives
Performance
Performance issue or suggestion for improvement
sota-implementations/
#4290
opened Sep 7, 2026 by
theap06
Collaborator
Loading…
[BugFix] Drop flaky B2/K16 tile from DreamerV3 block-GRU Triton
BugFix
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Integrations/torch_geometric
Integrations
Modules
#4257
opened Sep 7, 2026 by
YeonwooSung
Contributor
Loading…
4 of 6 tasks
[BugFix] Stop sharing Minari mission stacks between observation and next
BugFix
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Data/minari
Data
Data-related PR, will launch data-related jobs
Environments
Adds or modifies an environment wrapper
#4256
opened Sep 7, 2026 by
YeonwooSung
Contributor
Loading…
4 of 6 tasks
[BugFix] Do not invent vLLM prompt log-probabilities
BugFix
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Integrations/torch_geometric
Integrations
llm/
LLM-related PR, triggers LLM CI tests
Modules
#4255
opened Sep 7, 2026 by
YeonwooSung
Contributor
Loading…
7 tasks done
[Doc] Document multi-agent _step TensorDict layout
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
#4232
opened Sep 5, 2026 by
YeonwooSung
Contributor
Loading…
4 of 6 tasks
[Doc] Document LossModule and collector weight sharing
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Collectors
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Objectives
tutorials/
#4231
opened Sep 5, 2026 by
YeonwooSung
Contributor
Loading…
5 of 10 tasks
[Doc] Document checkpointing to resume RL training
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
#4230
opened Sep 5, 2026 by
YeonwooSung
Contributor
Loading…
4 of 6 tasks
[Doc] Document CatFrames with collectors and image replay buffers
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Transforms
#4229
opened Sep 5, 2026 by
YeonwooSung
Contributor
Loading…
4 of 6 tasks
[Performance] Add Triton backend to DreamerV3 RSSM
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Modules
Objectives
sota-implementations/
[Feature] Isaac Lab 3.x compatibility for wrapper
BugFix
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Environments/isaaclab
Environments
Adds or modifies an environment wrapper
Feature
New feature
#4194
opened Sep 1, 2026 by
theap06
Collaborator
Loading…
[Feature] Transformer Module
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Feature
New feature
Integrations/torch_geometric
Integrations
Modules
Transforms
#4193
opened Sep 1, 2026 by
theap06
Collaborator
Loading…
[Doc] Fold corrected TRL interoperability end-to-end section into trl_interop tutorial
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
tutorials/
#4161
opened Aug 26, 2026 by
coder-jayp
Contributor
Loading…
4 of 6 tasks
[CI] Retry flaky-test artifact downloads
CI
Has to do with CI setup (e.g. wheels & builds, tests...)
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
#4153
opened Aug 24, 2026 by
CoffeeDrivenCoder
Loading…
3 tasks done
[Feature] Add COMA multi-agent objective
CI
Has to do with CI setup (e.g. wheels & builds, tests...)
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Feature
New feature
Objectives
sota-implementations/
tutorials/
#4151
opened Aug 24, 2026 by
Iliamsou
Loading…
[Deprecation] Make DreamerV3 defaults canonical
Benchmarks
rl/benchmark changes
CI
Has to do with CI setup (e.g. wheels & builds, tests...)
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Deprecation
Documentation
Improvements or additions to documentation
Integrations/torch_geometric
Integrations
Modules
Objectives
sota-implementations/
[Performance] Improve geometric window sampling
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Performance
Performance issue or suggestion for improvement
ReplayBuffers
#4131
opened Aug 19, 2026 by
theap06
Collaborator
Loading…
[Feature] Add geometric trajectory window sampler
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Examples
Feature
New feature
ReplayBuffers
[Feature] Add FixedBatchedInference helper (AsyncEnvPool v2 item 3)
Benchmarks
rl/benchmark changes
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Documentation
Improvements or additions to documentation
Feature
New feature
tutorials/
user-facing
User-facing changes - only include in major releases
#4110
opened Aug 15, 2026 by
coder-jayp
Contributor
Loading…
5 of 6 tasks
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.