2 Releases published by 2 people

• ray-2.44.0 Ray-2.44.0

  published Mar 21, 2025

• ray-2.44.1 Ray-2.44.1

  published Mar 27, 2025

90 Pull requests merged by 33 people

• [CI] Add ability to configure release tests with a matrix

  #51721 merged Mar 27, 2025

• [docker] Update latest Docker dependencies for 2.44.1 release

  #51775 merged Mar 27, 2025

• [ci] Allow dependency failure for kuberay ci kickoff

  #51760 merged Mar 27, 2025

• [core] disable asan tests on premerge

  #51772 merged Mar 27, 2025

• [Chore][Autoscaler] Clarify disable_launch_config_check comment in StandardAutoscaler

  #51751 merged Mar 27, 2025

• [data] refine backpressure info on progress bar

  #51697 merged Mar 27, 2025

• [Data] Revisiting make_async_gen to address issues with concurrency control for sequences of varying lengths

  #51661 merged Mar 27, 2025

• [core][state] Fix false alarm in get_logs when a server chunk splits into multiple client chunks

  #51750 merged Mar 27, 2025

• Revert "[doc] Add hpu resource description in ray train related docs"

  #51754 merged Mar 27, 2025

• Make @edoakes the czar of _common/ dir for now

  #51753 merged Mar 27, 2025

• [Feat][Core/Dashboard] Convert ReportHead to subprocess module

  #51733 merged Mar 27, 2025

• [core] Fix windows build with no cython -Wno-shadow

  #51730 merged Mar 27, 2025

• [data] add getdaft to compiled versions

  #51723 merged Mar 27, 2025

• [serve] Remove RAY_SERVE_EAGERLY_START_REPLACEMENT_REPLICAS flag

  #51722 merged Mar 26, 2025

• Revert "[serve] Log rejected requests at router side (#51346)"

  #51698 merged Mar 26, 2025

• Run basic Python 3.13 tests

  #51688 merged Mar 26, 2025

• [ci] add misc and untested files in skipping

  #51715 merged Mar 26, 2025

• [doc] Add hpu resource description in ray train related docs

  #47241 merged Mar 26, 2025

• [core] test_job_isolation passes even when exceptions are thrown

  #51694 merged Mar 26, 2025

• [core][kuberay] Trigger kuberay release pipeline from rayci

  #51539 merged Mar 26, 2025

• [data] Integrate Ray Dataset with Daft Dataframe

  #51531 merged Mar 26, 2025

• [Feat][Core/Dashboard] Convert StateHead to subprocess module

  #51676 merged Mar 26, 2025

• [core] Fix all gcs variable shadowing

  #51704 merged Mar 26, 2025

• [Autoscaler] Update CoordinatorNodeProvider example

  #51293 merged Mar 26, 2025

• Fix Ray Client when 'uv run' runtime environment is used

  #51683 merged Mar 26, 2025

• [core] Fix all raylet variable shadowing

  #51689 merged Mar 26, 2025

• [docs] Update usage_lib.py guide link

  #51681 merged Mar 26, 2025

• [core][autoscaler][v2] do not removing nodes for upcoming resource requests

  #51570 merged Mar 26, 2025

• [CI] Update LLM dependencies list and make the uv compile test job hard fail

  #51693 merged Mar 26, 2025

• [core] Fix incorrect comment

  #51575 merged Mar 25, 2025

• [tests] Reassign dashboard tests to core team

  #51691 merged Mar 25, 2025

• [core] Introduce ConcurrentFlatMap and use for InMemoryStoreClient

  #50375 merged Mar 25, 2025

• [Serve.llm] fix loading model from remote storage and add docs

  #51617 merged Mar 25, 2025

• [core] Record dashboard metrics with oneshot

  #51627 merged Mar 25, 2025

• [Feat][Core/Dashboard] Convert JobHead to subprocess module

  #51553 merged Mar 25, 2025

• [core] Avoid resize in GetAndPinArgsForExecutor

  #51543 merged Mar 25, 2025

• [release] Fix perf metrics compare

  #51655 merged Mar 25, 2025

• [Data][LLM] trust remote code

  #51680 merged Mar 25, 2025

• [core] Fix all variable shadowing for core worker

  #51672 merged Mar 25, 2025

• [core] Threaded actors get stuck forever if they receive two exit signals

  #51582 merged Mar 25, 2025

• [core] [easy] Mark cgroup tests exclusive

  #51654 merged Mar 25, 2025

• [Data] Support async callable classes in flat_map()

  #51180 merged Mar 25, 2025

• [Data] fix RandomAccessDataset.multiget returning unexpected values for missing keys

  #44769 merged Mar 25, 2025

• [data] fix lance ut failed

  #51421 merged Mar 25, 2025

• [core] Correct the wording in the OnNodeDead logs to avoid confusion

  #51668 merged Mar 25, 2025

• [ci] add an always tag for cond testing

  #51662 merged Mar 25, 2025

• [deps] Use UV to compile LLM dependencies

  #51323 merged Mar 25, 2025

• [data] Update repartition on target_num_rows_per_block documentation

  #51433 merged Mar 25, 2025

• [data] Fix Databricks host URL handling in Ray Data

  #49926 merged Mar 25, 2025

• [serve] Remove RAY_SERVE_ENABLE_QUEUE_LENGTH_CACHE flag

  #51649 merged Mar 24, 2025

• [ray.llm] Refactor model download utilities

  #51604 merged Mar 24, 2025

• refactor replica _handle_errors_and_metrics

  #51644 merged Mar 24, 2025

• [ci] Enable Cgroup support in CI for core

  #51454 merged Mar 24, 2025

• [Test][KubeRay] Add doctest for RayCluster Quickstart doc

  #51249 merged Mar 24, 2025

• Skip multiplex metrics and proxy status code is error tests on windows

  #51645 merged Mar 24, 2025

• [serve] update deployment status docs

  #51610 merged Mar 24, 2025

• [RLlib] Make min/max env steps per evaluation sample call configurable for duration="auto".

  #51637 merged Mar 24, 2025

• Fix Ray Train release test

  #51624 merged Mar 24, 2025

• [serve] don't stop retrying replicas when a deployment is scaling back up from zero

  #51600 merged Mar 24, 2025

• [core] Fix test_threaded_actor flaky on mac

  #51602 merged Mar 24, 2025

• Fix syntax errors in Ray Tune example pbt_ppo_example.ipynb

  #51626 merged Mar 23, 2025

• [core] [easy] [noop] Add comments on client call

  #51614 merged Mar 22, 2025

• [Doc][KubeRay] Add a doc to explain why some worker Pods are not ready in RayService

  #51095 merged Mar 22, 2025

• [Feat][Core/Dashboard] Convert EventHead to subprocess module

  #51587 merged Mar 22, 2025

• [Core] Cover cpplint for /src/ray/core_worker (excluding transport)

  #51557 merged Mar 22, 2025

• [Docs][Core] Update system logs doc for dashboard subprocess module

  #50984 merged Mar 22, 2025

• [Feat][Core/Dashboard] Convert DataHead to subprocess module

  #51507 merged Mar 22, 2025

• Add TorchDataLoader to Train Benchmark

  #51456 merged Mar 22, 2025

• [core] [easy] [no-op] Fix rotation comment

  #51606 merged Mar 21, 2025

• [release-automation] Add option to add build tag when uploading wheels to pypi

  #51517 merged Mar 21, 2025

• [serve][tests] Add a timeout for resnet app image request

  #51569 merged Mar 21, 2025

• [serve][test] Change the response_time_s to response_time_ms

  #51566 merged Mar 21, 2025

• Fix broken doctest build

  #51594 merged Mar 21, 2025

• [Serve.llm] Add gen config related doc

  #51572 merged Mar 21, 2025

• [CI] Upgrade pytest-aiohttp to 1.1.0

  #51556 merged Mar 21, 2025

• [Data] Removing usages of the deprecated use_legacy_format param

  #51563 merged Mar 21, 2025

• [llm] ray.llm support custom accelerators

  #51359 merged Mar 21, 2025

• [Doc] Clarify the relation between 'uv run' and 'uv pip' support

  #51599 merged Mar 21, 2025

• [Data] Adding more ops to BlockColumnAccessor

  #51571 merged Mar 21, 2025

• [ray.data.llm] Propose log_input_column_names()

  #51441 merged Mar 21, 2025

• Move experimental and OOM tests to core builds

  #51525 merged Mar 21, 2025

• Add perf metrics for 2.44.0

  #51427 merged Mar 21, 2025

• [chore][core] Make the error message more actionable when Protobuf-generated code and custom_types.py are inconsistent

  #51568 merged Mar 21, 2025

• [core] Make testable stream redirection

  #51191 merged Mar 21, 2025

• [Feat][Core/Dashboard] Remove ReportEventService and replace with HTTP API

  #51555 merged Mar 21, 2025

• [docker] Update latest Docker dependencies for 2.44.0 release

  #51581 merged Mar 21, 2025

• [docker] Update latest Docker dependencies for 2.44.0 release

  #51580 merged Mar 21, 2025

• [Feat][Core/Dashboard] Add more shared properties to subprocess module and supports websocket handler returns normal HTTP response

  #51552 merged Mar 21, 2025

• [Feat][Core/Dashboard] Redirect child process stdout and stderr to dashboard_[module_name].err

  #51545 merged Mar 21, 2025

• Give better error message if 'uv run' is combined with incompatible plugins

  #51565 merged Mar 21, 2025

70 Pull requests opened by 35 people

• [WIP] [core] Rotate log monitor

  #51573 opened Mar 21, 2025

• Bump gradio from 3.50.2 to 5.22.0 in /python/requirements

  #51577 opened Mar 21, 2025

• Bump transformers from 4.30.1 to 4.48.0 in /doc/source/templates/testing/docker/03_serving_stable_diffusion

  #51597 opened Mar 21, 2025

• Bump vllm from 0.7.2 to 0.8.1 in /python

  #51603 opened Mar 21, 2025

• [docs] Feature: adopt llms.txt convention

  #51605 opened Mar 21, 2025

• Bump pytorch-lightning from 1.8.6 to 2.4.0 in /python

  #51607 opened Mar 21, 2025

• Bump flask-cors from 4.0.0 to 4.0.2 in /python

  #51609 opened Mar 21, 2025

• [serve] reorg

  #51611 opened Mar 21, 2025

• Bump torch from 2.0.1 to 2.4.0 in /doc/source/templates/testing/docker/03_serving_stable_diffusion

  #51612 opened Mar 21, 2025

• Bump mlflow from 2.9.2 to 2.20.3 in /python/requirements/ml

  #51613 opened Mar 22, 2025

• Bump gunicorn from 20.1.0 to 23.0.0 in /python

  #51615 opened Mar 22, 2025

• [core] grpc stub manager

  #51616 opened Mar 22, 2025

• correct the error msg for invalid env registering

  #51621 opened Mar 22, 2025

• [core] Large objects release test

  #51625 opened Mar 22, 2025

• [py_modules] Don't install the wheel package if it's already installed

  #51629 opened Mar 23, 2025

• [data] add hive catalog

  #51638 opened Mar 24, 2025

• [RLlib] MetricsLogger + Stats overhaul

  #51639 opened Mar 24, 2025

• [RLlib] APPO/IMPALA accelerate (vol 41): Log losses as detached+cpu tensors to avoid CUDA DDP synchronizations.

  #51640 opened Mar 24, 2025

• [core] split scheduler into smaller targets to improve build performance

  #51641 opened Mar 24, 2025

• [Core][Prototype] Prototype Code for Event Buffer

  #51648 opened Mar 24, 2025

• [core] Fix actor reconstruction that depends on plasma object

  #51653 opened Mar 24, 2025

• Add image datasets to ray train benchmark

  #51657 opened Mar 24, 2025

• fix status codes on http proxy

  #51658 opened Mar 25, 2025

• this commit adds use of specific python3.9 version for development se…

  #51663 opened Mar 25, 2025

• [VMware][WCP provider][Part 3/3] Architecture documentation & uts for vsphere wcp provider

  #51666 opened Mar 25, 2025

• [WIP] [core] Force compilation error on variable shadow

  #51669 opened Mar 25, 2025

• update to protbuf-28.2, absl-20240722, grpc-1.67 and patch for windows

  #51673 opened Mar 25, 2025

• Add uv to Docker image

  #51675 opened Mar 25, 2025

• windows dev setup

  #51678 opened Mar 25, 2025

• [Refactor][Core/Dashboard] Extract shared implementation from ServeHead and ServeAgent to a separate class

  #51682 opened Mar 25, 2025

• [docs] Tune toc

  #51684 opened Mar 25, 2025

• [core] add actor labels to export events

  #51687 opened Mar 25, 2025

• [core] Avoid task spec copy for in order actor task submission

  #51692 opened Mar 25, 2025

• Revert "[cg] Move default device logic into channel utils (#51305)"

  #51699 opened Mar 26, 2025

• [WIP] [core] Add worker process into application cgroup

  #51701 opened Mar 26, 2025

• [core][dashboard-agent] Fail fast if the dashboard agent fails to launch the HTTP server

  #51705 opened Mar 26, 2025

• Update `--labels` and add `--labels-from-file` options for Label Selector API

  #51706 opened Mar 26, 2025

• Add `label_selector` option to remote functions

  #51707 opened Mar 26, 2025

• [Docs][KubeRay] Add guide for writing KubeRay doctests

  #51708 opened Mar 26, 2025

• [Test][KubeRay] Add a deliberate failure test to ensure doctests fail on error

  #51709 opened Mar 26, 2025

• [data] Make sql cursor buffered

  #51712 opened Mar 26, 2025

• [ci] upgrade rayci version

  #51713 opened Mar 26, 2025

• [core] Lazily subscribe to node changes from workers

  #51718 opened Mar 26, 2025

• [Core] Native CPU affinity support for accelerators

  #51719 opened Mar 26, 2025

• [Data][LLM] Bump vLLM version to support new models

  #51726 opened Mar 26, 2025

• [WIP]

  #51727 opened Mar 26, 2025

• [train] differentiate between train v1 and v2 export data

  #51728 opened Mar 26, 2025

• [WIP] [core] Log rotation for dashboard

  #51729 opened Mar 27, 2025

• [core] Log rotate monitor

  #51731 opened Mar 27, 2025

• [core][cgraph] Fix illegal memory access of cgraph when used in PP

  #51734 opened Mar 27, 2025

• [core] Set actor creation task's `num_returns` to 0 instead of 1

  #51735 opened Mar 27, 2025

• [Chore][Core/Dashboard] Remove TrainHead's dependency on DataOrganizer

  #51739 opened Mar 27, 2025

• [Core] Runtime env working_dir validation #51380

  #51741 opened Mar 27, 2025

• [core] Fix GCS target compilation

  #51742 opened Mar 27, 2025

• [WIP] [core] Fix root variable shadow

  #51743 opened Mar 27, 2025

• [data] support new pyiceberg version

  #51744 opened Mar 27, 2025

• [WIP] [core] Avoid ray_common as dependency

  #51745 opened Mar 27, 2025

• [Test][Dashboard] Add API tests for MetricsHead module

  #51752 opened Mar 27, 2025

• [Test][KubeRay] Add doctest for RayJob Quickstart doc

  #51756 opened Mar 27, 2025

• [core] Remove unnecessary exporter depencencies

  #51763 opened Mar 27, 2025

• [Data] Remove lazy fixture

  #51764 opened Mar 27, 2025

• [core] debug

  #51765 opened Mar 27, 2025

• [core] Remove object store runner

  #51766 opened Mar 27, 2025

• export target details via applications rest api

  #51767 opened Mar 27, 2025

• [Serve] Unify request cancellation errors

  #51768 opened Mar 27, 2025

• [Data] Implement forceful releasing of actors upon shutdown of `StreamingExecutor`

  #51769 opened Mar 27, 2025

• [ci] mark jailed rllib tests as manual

  #51770 opened Mar 27, 2025

• [Core][Autoscaler] Update the Autoscaler Resource Requests Data Model for Scheduling for Label Selector

  #51771 opened Mar 27, 2025

• [rllib] apply buildifier to build file

  #51773 opened Mar 27, 2025

• [ci] add flaky tags for flaky test groups

  #51774 opened Mar 27, 2025

55 Issues closed by 18 people

• Release test rllib_multi_gpu_with_attention_learning_tests.aws failed

  #42605 closed Mar 27, 2025

• Release test rllib_learner_group_checkpointing_multinode.aws failed

  #44209 closed Mar 27, 2025

• Release test rllib_learning_tests_marwil_old_api_stack_tf.aws failed

  #44522 closed Mar 27, 2025

• Release test rllib_learning_tests_cql_old_api_stack_tf.aws failed

  #44552 closed Mar 27, 2025

• CI test linux://rllib:learning_tests_multi_agent_cartpole_appo_multi_cpu is flaky

  #46330 closed Mar 27, 2025

• CI test linux://rllib:learning_tests_cartpole_truncated_ppo is flaky

  #47646 closed Mar 27, 2025

• CI test linux://rllib:learning_tests_cartpole_truncated_ppo is flaky

  #47635 closed Mar 27, 2025

• CI test windows://python/ray/serve/tests:test_metrics is flaky

  #45843 closed Mar 27, 2025

• CI test linux://python/ray/tests:test_state_api_2 is flaky

  #51736 closed Mar 27, 2025

• CI test linux://python/ray/tests:test_asyncio_client_mode is flaky

  #51659 closed Mar 27, 2025

• CI test linux://python/ray/tests:test_asyncio is flaky

  #51660 closed Mar 27, 2025

• CI test linux://doc:doctest[train-gpu][gpu] is consistently_failing

  #51740 closed Mar 27, 2025

• AssertionError: Session name mismatch after restarting Ray cluster with cleanup steps

  #51737 closed Mar 27, 2025

• [Ray debugger] Ray dubbger does not work

  #51670 closed Mar 27, 2025

• 【bug】Ray.data.write_parquet will write twice when use fsspec local filesystem

  #49741 closed Mar 26, 2025

• [data] iter_batches needs streaming operation

  #49072 closed Mar 26, 2025

• [data] Documentation is formatted incorrectly

  #48974 closed Mar 26, 2025

• CI test linux://python/ray/tests:test_array_asan is consistently_failing

  #51714 closed Mar 26, 2025

• CI test linux://python/ray/serve/tests:test_max_replicas_per_node_with_stop_fully_then_start_behavior is consistently_failing

  #51716 closed Mar 26, 2025

• CI test linux://python/ray/serve/tests:test_multiplex_with_queue_len_cache_disabled is consistently_failing

  #48379 closed Mar 26, 2025

• [<Ray component: Core|RLlib|etc...>] why ? I use 'curl -v http://127.0.0.1:8268/api/version' return @@@? ,

  #51584 closed Mar 26, 2025

• [Ray Core] The node storing the actor will be kill unexpectedly when autoscaler is turned on

  #46172 closed Mar 26, 2025

• [Core] Problems with uv run and remote cluster

  #51368 closed Mar 26, 2025

• [core][autoscaler][v2] do not removing nodes for upcoming resource requests

  #51321 closed Mar 26, 2025

• CI test linux://rllib:learning_tests_multi_agent_pendulum_sac_multi_cpu is flaky

  #47264 closed Mar 26, 2025

• [data] async `flat_map`

  #50329 closed Mar 26, 2025

• [data] Async map_batches return empty result when execution_options.preserve_order = True

  #51188 closed Mar 26, 2025

• CI test linux://python/ray/dashboard:test_node is consistently_failing

  #51618 closed Mar 25, 2025

• CI test linux://python/ray/dashboard:test_dashboard is consistently_failing

  #44917 closed Mar 25, 2025

• CI test darwin://python/ray/tests:test_gcs_fault_tolerance is consistently_failing

  #43777 closed Mar 25, 2025

• CI test darwin://python/ray/tests:test_scheduling_performance is flaky

  #44238 closed Mar 25, 2025

• [Serve] Elastic Autoscaling Based on Cluster Resources with Customizable Scaling Logic

  #49151 closed Mar 25, 2025

• [Data] RandomAccessDataset.multiget return unexpected values for missing keys.

  #44768 closed Mar 25, 2025

• Release test training_ingest_benchmark-task=image_classification.skip_training failed

  #51622 closed Mar 25, 2025

• Release test training_ingest_benchmark-task=image_classification.skip_training.fault_tolerance failed

  #51623 closed Mar 25, 2025

• Release test training_ingest_benchmark-task=image_classification.skip_training_torch_dataloader failed

  #51633 closed Mar 25, 2025

• CI test windows://python/ray/tests:test_actor_client_mode is flaky

  #51651 closed Mar 25, 2025

• CI test linux://python/ray/data:test_huggingface is consistently_failing

  #44516 closed Mar 25, 2025

• [<Ray component: Data] - inconsistent URL handling in Ray's Databricks integration

  #49925 closed Mar 25, 2025

• [client] Documentation Python version behavior

  #45339 closed Mar 24, 2025

• Release test aws_cluster_launcher_minimal failed

  #51443 closed Mar 24, 2025

• Release test aws_cluster_launcher failed

  #51437 closed Mar 24, 2025

• [Serve] Serve no longer retries deployments after 3 failures

  #50710 closed Mar 24, 2025

• [Data/preprocessors] Allow preprocessors to be append operations

  #48133 closed Mar 24, 2025

• CI test darwin://python/ray/tests:test_task_metrics is flaky

  #48278 closed Mar 22, 2025

• [Core] Cover cpplint for `ray/core_worker` (excluding transport)

  #51510 closed Mar 22, 2025

• CI test linux://python/ray/serve/tests/unit:test_pow_2_replica_scheduler is flaky

  #48736 closed Mar 22, 2025

• [core] Place unit test alongside with the implementation

  #51152 closed Mar 21, 2025

• CI test linux://python/ray/data:test_parquet is flaky

  #48152 closed Mar 21, 2025

• CI test linux://python/ray/data:test_metadata_provider is flaky

  #51436 closed Mar 21, 2025

• [Core] API Reference: uv

  #51195 closed Mar 21, 2025

• CI test linux://python/ray/tests:test_object_spilling_2_debug_mode is consistently_failing

  #49143 closed Mar 21, 2025

• [core] Compiled Graphs has a dependence on pyarrow

  #51595 closed Mar 21, 2025

• Ray client connection timeout on ray.init

  #51591 closed Mar 21, 2025

• [Core] Failed to use uv

  #51196 closed Mar 21, 2025

46 Issues opened by 29 people

• [core][dashboard] don't use the first byte to determine whether a chunk succeeds or not in `get_log`

  #51762 opened Mar 27, 2025

• [Ray Serve] request timeout sec for grpc

  #51761 opened Mar 27, 2025

• Release test stress_test_state_api_scale.aws failed

  #51759 opened Mar 27, 2025

• [Data] Found Bug `add_column`

  #51758 opened Mar 27, 2025

• [Ray Data: Preprocessors] Support flattening vector features in concatenator

  #51757 opened Mar 27, 2025

• [Core] `uv sync` fails when running fork of Ray

  #51755 opened Mar 27, 2025

• CI test darwin://python/ray/tests:test_state_api_2 is consistently_failing

  #51749 opened Mar 27, 2025

• [RLLIB] PPO Gradient Removed for Value Estimation

  #51748 opened Mar 27, 2025

• [RLLIB] Offline Training

  #51747 opened Mar 27, 2025

• CI test darwin://python/ray/tests:test_state_api_log is consistently_failing

  #51746 opened Mar 27, 2025

• CI test linux://python/ray/tests:test_job is flaky

  #51738 opened Mar 27, 2025

• CI test linux://python/ray/air:test_resource_manager_placement_group is consistently_failing

  #51725 opened Mar 26, 2025

• [serve.llm][Feature request] Adding new models to a multi-gpu multi-model service would require the duplication of all the resources

  #51720 opened Mar 26, 2025

• [<Ray component: Core>,C++] Ray worker process number keep increasing if calling actor from workers

  #51711 opened Mar 26, 2025

• CI test darwin://python/ray/dashboard:test_cli_integration is flaky

  #51710 opened Mar 26, 2025

• Outdated docs

  #51703 opened Mar 26, 2025

• (P1) Update Autoscaler data model to pass label information by adding a labels field to the `ResourceRequest` message

  #51702 opened Mar 26, 2025

• DQN training with 2.44.0 with gymnasium.env and action mask giving error

  #51700 opened Mar 26, 2025

• [core] Ray status doesn't work for all ostream-printable objects

  #51695 opened Mar 25, 2025

• [core][gpu-objects] intra-process communication

  #51685 opened Mar 25, 2025

• AttributeError observed in sample_func when executed from a different python source file

  #51679 opened Mar 25, 2025

• [core/scheduler] replace :ray_common dep with sub-dependencies

  #51677 opened Mar 25, 2025

• [core] Add more compilation options and link options

  #51671 opened Mar 25, 2025

• [core] Ray status doesn't show source location where the error happens

  #51667 opened Mar 25, 2025

• [build] Avoid ODR issues

  #51647 opened Mar 24, 2025

• [Core] Expose `tags` parameter for tasks/actors to be propagated to metrics

  #51646 opened Mar 24, 2025

• [core][gpu-objects] Support streaming to overlap computation / communication

  #51643 opened Mar 24, 2025

• [core] Unify `CoreWorker::Exit` and `CoreWorker::Shutdown`

  #51642 opened Mar 24, 2025

• [<Ray component: Core|RLlib|etc...>] No action space and observation space in MASAC algorithm in the tune_example

  #51636 opened Mar 24, 2025

• [core/scheduler] Split giant ray core C++ target into small ones

  #51634 opened Mar 24, 2025

• [Serve] Ray Serve Autoscaling supports the configuration of custom-metrics and policy

  #51632 opened Mar 24, 2025

• RLlib new API stack false deprecation warning / MultiRLModuleSpec

  #51630 opened Mar 23, 2025

• [Core] Unable to build Ray wheel on Windows using Docker due to private image access issues

  #51628 opened Mar 23, 2025

• [RLlib] Incorrect error message for improper registering of custom env

  #51620 opened Mar 22, 2025

• [core] Split giant ray core C++ targets into small ones(plasma store)

  #51619 opened Mar 22, 2025

• [Cluster] Ray job submit/logs sporadically stops following logs

  #51601 opened Mar 21, 2025

• [Ray serve] StopAsyncIteration error thrown by ray when the client cancels the request

  #51598 opened Mar 21, 2025

• [CG, Core] Illegal memory access with Ray 2.44 and vLLM v1 pipeline parallelism

  #51596 opened Mar 21, 2025

• [cgraph] Support function nodes

  #51593 opened Mar 21, 2025

• [Cluster] Add uv to base images

  #51592 opened Mar 21, 2025

• [core] Combine multiple grpc connections into one

  #51590 opened Mar 21, 2025

• [core] Replace opencensus with opentelemetry (C++)

  #51589 opened Mar 21, 2025

• [Cluster] Split up monitor.log

  #51586 opened Mar 21, 2025

• [Cluster] Autoscaler frequently fails to scale down workers

  #51585 opened Mar 21, 2025

• [Ray Core] setting a large max_concurrency value for actor causes `Unhandled exception: N5boost10wrapexceptINS_6system12system_errorEEE. what(): thread: Resource temporarily unavailable [system:11]`

  #51583 opened Mar 21, 2025

• [CG, Core] Add Ascend NPU Support for RCCL and CG

  #51574 opened Mar 21, 2025

788 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [Core] Making Object Store Fallback Directory Configurable

  #51189 commented on Mar 27, 2025 • 13 new comments

• [Data] Update whether release tests use autoscaling

  #51562 commented on Mar 27, 2025 • 12 new comments

• [Compiled Graph] Enhance Compile Graph with Multi-Device Support

  #51032 commented on Mar 27, 2025 • 9 new comments

• [data] make random_sample() reproducible

  #51401 commented on Mar 26, 2025 • 8 new comments

• [LLM Batch] SGLang engine stage and processor

  #51409 commented on Mar 27, 2025 • 7 new comments

• [Core] Cover cpplint for ray/src/ray/common

  #51551 commented on Mar 26, 2025 • 5 new comments

• [core] support dir includes env for working dir

  #50066 commented on Mar 26, 2025 • 5 new comments

• [data] adding snowflake connectors

  #51429 commented on Mar 26, 2025 • 5 new comments

• [Grafana] Enable `includeAll` for Grafana cluster variable

  #51396 commented on Mar 27, 2025 • 5 new comments

• [Core][Bug fix] Trigger local task scheduling after deleting bundle.

  #51125 commented on Mar 26, 2025 • 5 new comments

• [doc] minor/patch version update

  #48626 commented on Mar 26, 2025 • 2 new comments

• [Autoscaler][V2] Check IM instance_status before terminating nodes

  #50707 commented on Mar 26, 2025 • 2 new comments

• [core][compiled graphs] Support reduce scatter and all gather collective for GPU communicator in compiled graph

  #50624 commented on Mar 26, 2025 • 2 new comments

• [core] Always create a default executor

  #51058 commented on Mar 26, 2025 • 1 new comment

• [core] Use cord for sending objects

  #51397 commented on Mar 26, 2025 • 1 new comment

• [data] Iceberg datasource read with pyiceberg 0.9 fix

  #51453 commented on Mar 26, 2025 • 1 new comment

• [core] Implement utils class to setup and cleanup cgroup folder

  #49941 commented on Mar 26, 2025 • 1 new comment

• Update rllib-env.rst

  #46750 commented on Mar 26, 2025 • 0 new comments

• [Core]Support Merge code search path from env variable

  #46771 commented on Mar 26, 2025 • 0 new comments

• Add docs link to Serve page of Ray Dashboard

  #46812 commented on Mar 26, 2025 • 0 new comments

• Introducing StaleTaskError

  #46705 commented on Mar 26, 2025 • 0 new comments

• [Core] If possible, force flush the trace when the worker ends.

  #46654 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Place the submit job on a separate page

  #46613 commented on Mar 26, 2025 • 0 new comments

• [ADAG] Fix DAG input

  #46604 commented on Mar 26, 2025 • 0 new comments

• Fix mlflow artifact logging

  #46570 commented on Mar 26, 2025 • 0 new comments

• [Data] Make the seed take effect in Dataset.random_sample()

  #46088 commented on Mar 26, 2025 • 0 new comments

• Updated LogVirtualView component removed react window

  #46835 commented on Mar 26, 2025 • 0 new comments

• Verification to move PyG data to device

  #46839 commented on Mar 26, 2025 • 0 new comments

• Add generic item support for queue

  #46849 commented on Mar 26, 2025 • 0 new comments

• [POC] A Reactor style GCS. #1: GcsNodeManager

  #46891 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Deprecate algo config (python) dicts; must be `AlgorithmConfig` objects.

  #46896 commented on Mar 26, 2025 • 0 new comments

• [Dashboard]. Map logic resource data to row node

  #46914 commented on Mar 26, 2025 • 0 new comments

• Added support for multiple callbacks for GcsSubscriber

  #46958 commented on Mar 26, 2025 • 0 new comments

• [ci][core] GCS FT Chaos test

  #46996 commented on Mar 26, 2025 • 0 new comments

• [core][dashboard] Change the StateDataSourceClient from using gRPC stub -> NewGcsClient.

  #47056 commented on Mar 26, 2025 • 0 new comments

• [core] GcsPublisher bindings

  #47062 commented on Mar 26, 2025 • 0 new comments

• Add Runhouse to Ecosystem

  #47150 commented on Mar 26, 2025 • 0 new comments

• Revert "[doc]Make vllm example works with latest vllm version"

  #46094 commented on Mar 26, 2025 • 0 new comments

• Add PyFlyt waypoints example to documentation

  #46145 commented on Mar 26, 2025 • 0 new comments

• [ADAG] Detect if ADAG is at capacity for execution

  #46158 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Add cleanup of `job_table` in `delete_job`

  #46173 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Add GPU component usage

  #46188 commented on Mar 26, 2025 • 0 new comments

• [core] add ray.util.concurrent.futures.RayExecutor

  #46249 commented on Mar 26, 2025 • 0 new comments

• [WIP] CI: jemalloc & mimalloc

  #46271 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Display accelerators info on demand, add Huawei Ascend NPU monitoring.

  #46287 commented on Mar 26, 2025 • 0 new comments

• [data] fix np.array crash the allocate mem error when souce include short an…

  #46298 commented on Mar 26, 2025 • 0 new comments

• Update py_modules.py AttributeError: module has no attribute '__path__'

  #46302 commented on Mar 26, 2025 • 0 new comments

• [Doc] Update directory path for installation

  #46318 commented on Mar 26, 2025 • 0 new comments

• Enable RAY_DATA_ENABLE_TENSOR_EXTENSION_CASTING environment variable

  #46344 commented on Mar 26, 2025 • 0 new comments

• fixed a typo in ValueError message for contains_tensor

  #46348 commented on Mar 27, 2025 • 0 new comments

• [Docker] Upgrade base deps docker python env to 3.9.7

  #46353 commented on Mar 26, 2025 • 0 new comments

• [test] cpp20

  #46380 commented on Mar 26, 2025 • 0 new comments

• [Core] Add ray-start option 'session-name'

  #46404 commented on Mar 26, 2025 • 0 new comments

• avoid merge errors when blocks contain different type in DelegatingBl…

  #46407 commented on Mar 26, 2025 • 0 new comments

• [Core] Use real CPU count available to a Ray process

  #46424 commented on Mar 26, 2025 • 0 new comments

• fix performance bug in arrow to numpy transform

  #46433 commented on Mar 26, 2025 • 0 new comments

• [Doc][KubeRay] Add KubeRay image resize example to Ray doc page

  #46447 commented on Mar 26, 2025 • 0 new comments

• python/ray/autoscaler/gcp/*.yaml: change scheduling from dict to list

  #46500 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Optimize rnn_sequencing performance

  #46502 commented on Mar 26, 2025 • 0 new comments

• [autoscaler][aws] Fix replace cloudwatch alarm config

  #46537 commented on Mar 26, 2025 • 0 new comments

• [Core]Fix the issue of actor tasks hanging during resubmission

  #46539 commented on Mar 27, 2025 • 0 new comments

• [Data] Enable streaming json read

  #46550 commented on Mar 26, 2025 • 0 new comments

• [Jobs] Making sure `JobManager` retries `JobSupervisor.ping` before declaring job as failed

  #47166 commented on Mar 26, 2025 • 0 new comments

• [ADAG]Enable NPU (hccl) communication for CG

  #47658 commented on Mar 26, 2025 • 0 new comments

• [wip] revive zero copy torch tensor serialization

  #47665 commented on Mar 26, 2025 • 0 new comments

• [Data] Fix parallelism deriving heuristic to ensure parallelism stays w/in min/max bounds

  #47695 commented on Mar 26, 2025 • 0 new comments

• [RLlib] PPO enhancement: Send samples as refs to n Learners (speedup for multi-node/multi-GPU learning).

  #47707 commented on Mar 26, 2025 • 0 new comments

• Convert export events to proto and flush from background thread

  #47713 commented on Mar 26, 2025 • 0 new comments

• [core]Make GCS InternalKV workload configurable to the Policy.

  #47736 commented on Mar 26, 2025 • 0 new comments

• [Docs][hotfix] Correct the desc of nums of blocks

  #47741 commented on Mar 26, 2025 • 0 new comments

• [core][aDAG] Fix cpu tensor is automatically converted to gpu tensor

  #47742 commented on Mar 26, 2025 • 0 new comments

• Getinternalconfig and ioctx

  #47756 commented on Mar 26, 2025 • 0 new comments

• [Docs] Update map_reduce.ipynb chunk_size

  #47766 commented on Mar 26, 2025 • 0 new comments

• [RayCluster] Introduce how to run ray remote job with ray client （#47…

  #47771 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix MeanStdFilter: Do not accumulate `num_pushes` for RunningStats when merging connector states.

  #47794 commented on Mar 26, 2025 • 0 new comments

• Update vllm_openai_example.py for compatibility with latest vllm

  #47835 commented on Mar 26, 2025 • 0 new comments

• Add new serve autoscaling parameter `scaling_function`

  #47837 commented on Mar 26, 2025 • 0 new comments

• [observability][export-api] Write TrainRun events

  #47888 commented on Mar 26, 2025 • 0 new comments

• Adding image_uri to docstring of runtime_env

  #47905 commented on Mar 26, 2025 • 0 new comments

• [Docs] Update Volcano Integration with The New Flag

  #47911 commented on Mar 26, 2025 • 0 new comments

• [ADAG] Fix none output.

  #47918 commented on Mar 26, 2025 • 0 new comments

• [RLlib; Offline RL] Enable GPU and multi-GPU training for offline algorithms.

  #47929 commented on Mar 26, 2025 • 0 new comments

• [Azure][Cluster] Check tags when provided

  #47941 commented on Mar 26, 2025 • 0 new comments

• [autoscalerv2] use replicas in workerGroupSpecs as current workers number when initialize scale request to fix scale up target is wrong

  #47967 commented on Mar 26, 2025 • 0 new comments

• (WIP) [ADAG] Support dag.experimental_compile(_custom_nccl_group= nccl_group) in aDAG

  #47987 commented on Mar 26, 2025 • 0 new comments

• [data] add backpressure reason

  #48009 commented on Mar 26, 2025 • 0 new comments

• fix: WandbLogger crashing silently on a FileNotFoundError

  #50308 commented on Mar 26, 2025 • 0 new comments

• [core][dashboard] Make updates to DataSource.(node_workers|core_worker_stats) on delta.

  #47186 commented on Mar 26, 2025 • 0 new comments

• add stop and delete button for jobs that are of submission type

  #47189 commented on Mar 26, 2025 • 0 new comments

• [core] Make preloading Jemalloc configurable for worker

  #47243 commented on Mar 26, 2025 • 0 new comments

• Add tensorflow support to numpy_to_tensor connector

  #47246 commented on Mar 26, 2025 • 0 new comments

• [bazel] move python rules up

  #47260 commented on Mar 26, 2025 • 0 new comments

• [core] Decouple create worker vs pop worker request.

  #47268 commented on Mar 26, 2025 • 0 new comments

• [data] Fixed pyarrow error when the writer receives empty table

  #47270 commented on Mar 26, 2025 • 0 new comments

• [RLlib; docs] New API stack docs: Add `ConnectorV2` documentation

  #47278 commented on Mar 26, 2025 • 0 new comments

• idempotent replies by seq_no for sequential actors.

  #47314 commented on Mar 26, 2025 • 0 new comments

• [Core][aDAG] Remove busy waiting semaphore acquire in linux

  #47322 commented on Mar 26, 2025 • 0 new comments

• [todo] Migrate redis kv get sync

  #47348 commented on Mar 26, 2025 • 0 new comments

• Remove unnecessary string literal splits

  #47360 commented on Mar 26, 2025 • 0 new comments

• Return multiple best trials

  #47381 commented on Mar 26, 2025 • 0 new comments

• [PoC] Dashboard with Heads as Actors.

  #47414 commented on Mar 26, 2025 • 0 new comments

• [Core] Refine accelerator resource assessment for better node selection

  #47443 commented on Mar 26, 2025 • 0 new comments

• [core][dashboard] make a flamegraph on event loop lag.

  #47491 commented on Mar 26, 2025 • 0 new comments

• Improvements and Artificial Intelligence-based Improvements for Ray Cross-Language Functionality Testing

  #47499 commented on Mar 26, 2025 • 0 new comments

• Enhancements to Ray Cross-Language Testing Script: Automated Error Detection, Data Input Checking, and System Efficiency Enhancement using Artificial Intelligence

  #47558 commented on Mar 26, 2025 • 0 new comments

• [RLlib] add `SingleAgentRLModuleSpec` alias to `RLModuleSpec`

  #47560 commented on Mar 26, 2025 • 0 new comments

• [Doc][KubeRay] Document Fields that Will Not Trigger Downtime in RayService

  #47561 commented on Mar 26, 2025 • 0 new comments

• uint8_t* data ptr not used.

  #47565 commented on Mar 26, 2025 • 0 new comments

• [Do not merge] Run release tests for export API

  #47568 commented on Mar 26, 2025 • 0 new comments

• [Core][StreamingGenerator] Fix ray.get streaming object hang after node dead.

  #47583 commented on Mar 26, 2025 • 0 new comments

• [RLlib|New API|Inconsistency] LSTM Encoder lacks the output Linear, but stated in the docstring (#47625)

  #47626 commented on Mar 26, 2025 • 0 new comments

• [RPC] Added appropriate keep-alive configuration for Ray's internal RPCs

  #44612 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix invalid call of action_sampler_fn to support Keras 3

  #44700 commented on Mar 26, 2025 • 0 new comments

• Update prometheus-grafana.md and add grafana support allowed_origins

  #44701 commented on Mar 26, 2025 • 0 new comments

• Deflake test_threaded_actor

  #44709 commented on Mar 26, 2025 • 0 new comments

• WIP: Futex

  #44724 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Make RLlib learner support custom resources

  #44732 commented on Mar 26, 2025 • 0 new comments

• [core] object store data transfer zstd

  #44755 commented on Mar 26, 2025 • 0 new comments

• [DO NOT SUBMIT] debug sleep macos

  #44759 commented on Mar 26, 2025 • 0 new comments

• [core] Add 5s timeout to the log and err subscriber polls.

  #44761 commented on Mar 26, 2025 • 0 new comments

• [Core] Profile Ray start

  #44818 commented on Mar 26, 2025 • 0 new comments

• Add Ray train dashboard head module with mock data

  #44819 commented on Mar 27, 2025 • 0 new comments

• [WIP] Fixes the streaming generator hang on conn break

  #44838 commented on Mar 26, 2025 • 0 new comments

• [DO NOT SUBMIT] Pr 44234

  #44839 commented on Mar 26, 2025 • 0 new comments

• upgrade node to v20, latest LTS

  #44860 commented on Mar 26, 2025 • 0 new comments

• [serve] update long running release tests

  #44915 commented on Mar 26, 2025 • 0 new comments

• [ci][core] Add -flto and -fwhole-program-vtables

  #44919 commented on Mar 26, 2025 • 0 new comments

• [core] add ray.util.concurrent.futures.RayExecutor

  #44922 commented on Mar 26, 2025 • 0 new comments

• Debug tune repro

  #44936 commented on Mar 26, 2025 • 0 new comments

• [ci] Adds promethesus latencies for ray_dashboard_api_requests_duration_seconds_bucket for the Ray Core Tests.

  #44944 commented on Mar 26, 2025 • 0 new comments

• [WIP] Experimenting with streaming

  #44959 commented on Mar 26, 2025 • 0 new comments

• [proof-of-concept][dashboard] One event loop per module

  #44964 commented on Mar 26, 2025 • 0 new comments

• [Data] Support task reassignment in actor_pool_map_operator to improv…

  #44968 commented on Mar 26, 2025 • 0 new comments

• add more execution and iteration metrics to prometheus

  #44971 commented on Mar 26, 2025 • 0 new comments

• blind try on ubuntu upgrade ..

  #45427 commented on Mar 26, 2025 • 0 new comments

• [Hack] Hack the pickle to make relpath in SCRIPT_MODE working dir

  #43804 commented on Mar 26, 2025 • 0 new comments

• [WIP] In Driver CoreWorkerProcess, shutdown with the Exit method

  #43833 commented on Mar 26, 2025 • 0 new comments

• [flakey] Deflakey `darwin://python/ray/tests:test_gcs_fault_tolerance`

  #43922 commented on Mar 26, 2025 • 0 new comments

• [Core] Remove external storage upon sigterm for ray start

  #43941 commented on Mar 26, 2025 • 0 new comments

• remove flaky marker from test

  #44033 commented on Mar 26, 2025 • 0 new comments

• Fix for issue #43411 (BaseException error)

  #44038 commented on Mar 26, 2025 • 0 new comments

• [misc] Reformat train/tune BUILD files

  #44151 commented on Mar 26, 2025 • 0 new comments

• [misc] Reformat RLLib BUILD files

  #44153 commented on Mar 26, 2025 • 0 new comments

• [Jobs] [Dashboard] Changing cluster address resolution in get_address_for_submission_client

  #44186 commented on Mar 26, 2025 • 0 new comments

• [gRPC] Adding retry policies for all gRPC clients

  #44234 commented on Mar 26, 2025 • 0 new comments

• Ray IPv6 support

  #44252 commented on Mar 26, 2025 • 0 new comments

• [CI] Update kind version if it doesn't match pinned version

  #44268 commented on Mar 26, 2025 • 0 new comments

• Debug reference_count

  #44271 commented on Mar 26, 2025 • 0 new comments

• Deflake test

  #44333 commented on Mar 26, 2025 • 0 new comments

• Retry on stream rpc lost

  #44358 commented on Mar 26, 2025 • 0 new comments

• test

  #44377 commented on Mar 26, 2025 • 0 new comments

• [WIP] Offload execution of sync methods to event-loop's default executor

  #44406 commented on Mar 26, 2025 • 0 new comments

• change naming to intel gaudi habana for ray train example

  #44412 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Cleanup `examples` folder 02: Add shared value function example script for MultiAgentRLModule.

  #44421 commented on Mar 26, 2025 • 0 new comments

• Deflake test with long sleep

  #44433 commented on Mar 26, 2025 • 0 new comments

• Update bert.ipynb

  #44455 commented on Mar 26, 2025 • 0 new comments

• Remove SimpleImageViewer from EnvRunnerV2

  #44466 commented on Mar 26, 2025 • 0 new comments

• [data] add better support for list-typed fields when using `write_bigquery`

  #44564 commented on Mar 26, 2025 • 0 new comments

• [RLLib] Fix action masking example

  #44565 commented on Mar 26, 2025 • 0 new comments

• [RLlib] DreamerV3 on PyTorch.

  #45463 commented on Mar 26, 2025 • 0 new comments

• [train] Update Torch default timeout_s to use Torch's default timeout

  #45501 commented on Mar 26, 2025 • 0 new comments

• Create a singleton io context and thread, and standalone gcs client on it.

  #45524 commented on Mar 26, 2025 • 0 new comments

• [RLlib] fix VTrace in impala_tf_policy to support Keras 3

  #45562 commented on Mar 26, 2025 • 0 new comments

• [WIP] [Core] Support of reading Working dir from HUAWEI Object Storage Service (OBS)

  #45577 commented on Mar 26, 2025 • 0 new comments

• [core] Eagerly kill idle workers on job finish.

  #45633 commented on Mar 26, 2025 • 0 new comments

• [core][1/2] Add SubscribeAllActors to GcsClient.

  #45637 commented on Mar 26, 2025 • 0 new comments

• [core][2/2] Kill worker on root detached actor died.

  #45638 commented on Mar 26, 2025 • 0 new comments

• [WIP][Jobs] Revisit Job Agent to run Job Supervisors in-process

  #45664 commented on Mar 26, 2025 • 0 new comments

• [Core] Add warning when uploading large working dirs

  #45818 commented on Mar 26, 2025 • 0 new comments

• [WIP] Benchmark data shuffle

  #45847 commented on Mar 26, 2025 • 0 new comments

• Improve code snippet in docs to set up `ray[serve]` gRPC service

  #45862 commented on Mar 26, 2025 • 0 new comments

• MADDPG framework should be TensorFlow

  #45863 commented on Mar 26, 2025 • 0 new comments

• Enable setting OS disk size in Azure

  #45867 commented on Mar 26, 2025 • 0 new comments

• Adds new working dir upload protocol PLASMA, and use it in job submission.

  #45880 commented on Mar 26, 2025 • 0 new comments

• [spark] Fix nvidia-smi hanging issue

  #45896 commented on Mar 26, 2025 • 0 new comments

• Fix ax_client.create_experiment call

  #45902 commented on Mar 26, 2025 • 0 new comments

• Fix malformed `temp_dir` path when connecting Windows workers to cluster with Linux head

  #45930 commented on Mar 26, 2025 • 0 new comments

• [URL] Change the absolute path to a relative path to solve the ingres…

  #45933 commented on Mar 26, 2025 • 0 new comments

• [Data] Remove gaps between tasks in ray data.

  #45935 commented on Mar 26, 2025 • 0 new comments

• [Serve] Group `DeploymentHandle` autoscaling metrics pushes by process

  #45957 commented on Mar 26, 2025 • 0 new comments

• enable easy logging of images to tensorboard

  #46068 commented on Mar 26, 2025 • 0 new comments

• [RLlib] - `"Synchronized"` sampling for multi-agent buffers.

  #46083 commented on Mar 26, 2025 • 0 new comments

• [WIP] poc / hack relpath

  #45003 commented on Mar 26, 2025 • 0 new comments

• [WIP] add env var to enable debug

  #45009 commented on Mar 26, 2025 • 0 new comments

• RuntimeContext support get actor namespace

  #45025 commented on Mar 26, 2025 • 0 new comments

• Add roundtrip (ping-pong) microbenchmarks for accelerated DAG channels

  #45064 commented on Mar 26, 2025 • 0 new comments

• [wip][train][tune] handle s3fs permissions

  #45100 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Revisited `JobManager` log fetching infra to avoid blocking the event-loop

  #45117 commented on Mar 26, 2025 • 0 new comments

• [Jobs] Revisit Ray Job execution and monitoring

  #45120 commented on Mar 26, 2025 • 0 new comments

• [RLlib; Tune] Fix default behavior of default tune `CLIReporter` (based on `Algorithm._progress_metrics`).

  #45122 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix async (multiprocessing) gymnasium vector envs in `SingleAgentEnvRunner`.

  #45144 commented on Mar 26, 2025 • 0 new comments

• [RFC] Splitted Dashboard Heads.

  #45175 commented on Mar 26, 2025 • 0 new comments

• Add descriptive error message when deployment name not found

  #45181 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Removes ray.rpc.ReportEventService and Dashboard head as gRPC server.

  #45219 commented on Mar 26, 2025 • 0 new comments

• [Core] Improve logging during accelerator auto-detection

  #45240 commented on Mar 26, 2025 • 0 new comments

• [core] Change all object_size to uint64_t and use 0 for unknown. Also adds a method `ray.experimental.get_local_object_locations`

  #45247 commented on Mar 26, 2025 • 0 new comments

• grid_search resolution code optimization

  #45267 commented on Mar 26, 2025 • 0 new comments

• [POC][core] GcsClient async binding, aka remove PythonGcsClient.

  #45289 commented on Mar 26, 2025 • 0 new comments

• [Data] add reset pandas index when merge sorted blocks

  #45326 commented on Mar 26, 2025 • 0 new comments

• add links to eks site for neuron examples

  #45341 commented on Mar 26, 2025 • 0 new comments

• Modify Spark on Ray to support Pex and other virtualenvs + direct scr…

  #45354 commented on Mar 26, 2025 • 0 new comments

• [WIP] Reopen cpp test on mac

  #45374 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Enhance callbacks test case for EnvRunners; Add (optional) explicit `enable_multi_agent` setting to AlgorithmConfig.

  #45385 commented on Mar 26, 2025 • 0 new comments

• [serve] allow build_serve_application to happen in parallel

  #45394 commented on Mar 26, 2025 • 0 new comments

• add ray debugger references to ray docs

  #45414 commented on Mar 26, 2025 • 0 new comments

• [Data] Allow configuration of MAX_IMAGE_PIXELS in ImageDatasource

  #45415 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Hide GPU and GRAM columns from clusters and actors table if there are 0 rows with GPUs

  #50338 commented on Mar 26, 2025 • 0 new comments

• [core][1/N] Set gRPC deadline to ReportOCMetrics RPC

  #50370 commented on Mar 26, 2025 • 0 new comments

• [data] add ClickHouse sink

  #50377 commented on Mar 27, 2025 • 0 new comments

• [core] Move `overload remote` for actors

  #50412 commented on Mar 26, 2025 • 0 new comments

• [Autoscaler][V2] Use running node instances to rate-limit upscaling

  #50414 commented on Mar 26, 2025 • 0 new comments

• [tune] Remove loguniform's base

  #50415 commented on Mar 26, 2025 • 0 new comments

• [core][cgraph] Support individual submit_timeout

  #50424 commented on Mar 26, 2025 • 0 new comments

• [core] add RAY_IGNORE_VERSION_MISMATCH when ray start --address

  #50513 commented on Mar 26, 2025 • 0 new comments

• Revert "[core][cgraph] Rework DagRef Destruction (#49818)"

  #50529 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Enable spliting and zero padding of Dict observation

  #50589 commented on Mar 26, 2025 • 0 new comments

• [Core] Split stats_metric into smaller targets to improve build performance

  #50595 commented on Mar 27, 2025 • 0 new comments

• [chore] Delete unused build.sh

  #50649 commented on Mar 26, 2025 • 0 new comments

• [doc][core] Fix ray generator code example

  #50655 commented on Mar 26, 2025 • 0 new comments

• [WIP / try out] Use UV for Python 3.13 tests

  #50669 commented on Mar 26, 2025 • 0 new comments

• [core] Cover cpplint for ray/src/ray/stats

  #50678 commented on Mar 26, 2025 • 0 new comments

• Move `pydantic_compat` from `_private` to `_common`

  #50683 commented on Mar 26, 2025 • 0 new comments

• [core] [wip attempt] StatusOr union construction sometimes breaks windows build

  #50761 commented on Mar 26, 2025 • 0 new comments

• [WIP] Ray Collective Communication Lib Support HCCL Backend

  #50790 commented on Mar 26, 2025 • 0 new comments

• [docs] add missing step to install KubeRay in gke-gcs-bucket.md

  #50811 commented on Mar 26, 2025 • 0 new comments

• [WIP] Rebasing materialized dataset on iterator back-pressure is active upon materialization

  #50880 commented on Mar 26, 2025 • 0 new comments

• [CI] Enable pretty-format-java pre-commit hook

  #50957 commented on Mar 26, 2025 • 0 new comments

• [wip] add object detection notebooks

  #50965 commented on Mar 27, 2025 • 0 new comments

• [doc][kuberay]: add `kubectl ray get node` example

  #51271 commented on Mar 26, 2025 • 0 new comments

• [WIP] Adding `transform` utility to `Operator`

  #48620 commented on Mar 26, 2025 • 0 new comments

• [RFC][dashboard] Use aiohttp client for inter dependencies.

  #49932 commented on Mar 26, 2025 • 0 new comments

• [core] minor optimization for JoinPaths

  #49946 commented on Mar 26, 2025 • 0 new comments

• adding distributional critic example

  #49949 commented on Mar 26, 2025 • 0 new comments

• [RLlib; Offline] - Add single learner gpu training with preloading in `OfflinePreLearner`.

  #49960 commented on Mar 26, 2025 • 0 new comments

• Explicit comm

  #49979 commented on Mar 26, 2025 • 0 new comments

• Add Semi-Random Weighting to AutoScaler Node Scheduler

  #49983 commented on Mar 26, 2025 • 0 new comments

• [kuberay] fix deserialisation of custom resources in autoscaler config

  #49993 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Remove the dashboard grpc server.

  #50021 commented on Mar 26, 2025 • 0 new comments

• [core] Thread-safe gcs node manager

  #50024 commented on Mar 26, 2025 • 0 new comments

• [Core][Doc] Add support for Cambricon MLU

  #50026 commented on Mar 26, 2025 • 0 new comments

• [Train] Add Cambricon MLU support to Ray Train

  #50028 commented on Mar 26, 2025 • 0 new comments

• [WIP] Move execution loop to the same thread as the constructor of an actor

  #50032 commented on Mar 26, 2025 • 0 new comments

• [RLlib] LearnerConnector pipeline speedup.

  #50035 commented on Mar 26, 2025 • 0 new comments

• Add Cloud Logging example for Ray on GKE

  #50060 commented on Mar 26, 2025 • 0 new comments

• Update multi-agent-envs.rst

  #50075 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Make `config.episodes_to_numpy` False by default.

  #50077 commented on Mar 26, 2025 • 0 new comments

• tsan

  #50105 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Remove DataSource.ndoes listeners in StateHead with get_all_node_info.

  #50122 commented on Mar 26, 2025 • 0 new comments

• [core][collective] Avoid creation of `gloo_queue` in race condition

  #50132 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Move record_dashboard_metrics from MetricsHead to DashboardHead, remove .metrics property and convert MetricsHead.

  #50133 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Use cloudpickle to pickle SubprocessModule classes, and convert ServeHead.

  #50153 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Disable callbacks callable check for new api stack

  #50157 commented on Mar 26, 2025 • 0 new comments

• [dashboard] Actor and node head

  #50159 commented on Mar 26, 2025 • 0 new comments

• [core] unblocking macos tests by pinning aiohappyeyeballs to version 2.4.8

  #51288 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Fix timezones to deal with daylight savings

  #51314 commented on Mar 26, 2025 • 0 new comments

• [CI] Replace `black` with `ruff format`

  #51332 commented on Mar 26, 2025 • 0 new comments

• Deflake

  #51338 commented on Mar 26, 2025 • 0 new comments

• [RLlib] - Add state syncing to EnvRunner sample call in APPO.

  #51343 commented on Mar 26, 2025 • 0 new comments

• [Debugger] Random pick ray debugger port from range

  #51344 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Support reporting AMD GPU usage

  #51345 commented on Mar 27, 2025 • 0 new comments

• [RLlib] Throw better error if catalog can not be created for Atari environments, help atari users

  #51371 commented on Mar 26, 2025 • 0 new comments

• [core] Create small objects release test

  #51382 commented on Mar 26, 2025 • 0 new comments

• [core] Remove the unnecessary key `ActorID` of `concurrency_groups_cache_` in TaskReceiver

  #51403 commented on Mar 26, 2025 • 0 new comments

• [data] Implement Spark-like accumulators for Ray Data

  #51404 commented on Mar 26, 2025 • 0 new comments

• Upgrading Arrow dependency to latest stable version

  #51440 commented on Mar 26, 2025 • 0 new comments

• expose ObjectRef from DeploymentResponse

  #51444 commented on Mar 26, 2025 • 0 new comments

• [Data] Add environment variable support for Ray Data execution callbacks.

  #51449 commented on Mar 26, 2025 • 0 new comments

• [serve] move serve image_uri tests to serve CI

  #51451 commented on Mar 26, 2025 • 0 new comments

• [WIP][core][gpu-objects] CollectiveGroupManager

  #51460 commented on Mar 26, 2025 • 0 new comments

• Unify `_private/log.py` and `_private/ray_logging`

  #51461 commented on Mar 26, 2025 • 0 new comments

• [core] upgrading macos CI python 3.9 -> 3.9.2 to enable numpy serialization warnings

  #51462 commented on Mar 26, 2025 • 0 new comments

• [ray.serve.llm] Support vLLM v1

  #51490 commented on Mar 26, 2025 • 0 new comments

• [RLlib|Tune|Train] ValueError: Could not recover from checkpoint as it does not exist anymore

  #51515 commented on Mar 26, 2025 • 0 new comments

• Avoid len(), which causes static batch sizes on export.

  #51520 commented on Mar 26, 2025 • 0 new comments

• Add perf metrics for 2.44.1

  #51535 commented on Mar 26, 2025 • 0 new comments

• Update observability.md

  #51567 commented on Mar 26, 2025 • 0 new comments

• [WIP] Try upgrade cython to 3.1

  #50972 commented on Mar 26, 2025 • 0 new comments

• fix restore BUG "RuntimeError: Expected scalars to be on CPU, got cud…

  #50983 commented on Mar 26, 2025 • 0 new comments

• Fix editorconfig option name

  #50993 commented on Mar 26, 2025 • 0 new comments

• Suppress type error

  #50994 commented on Mar 26, 2025 • 0 new comments

• Improvements to General Debugging guide

  #51004 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Schedule `AggregatorActors` via `PlacementGroupSchedulingStrategy` into Learner bundles.

  #51017 commented on Mar 26, 2025 • 0 new comments

• [doc] add jax example

  #51040 commented on Mar 26, 2025 • 0 new comments

• [WIP][core][compiled graphs] Supporting allreduce on tuple of tensors

  #51047 commented on Mar 26, 2025 • 0 new comments

• [Misc]cupy.cuda.nccl.get_unique_id() generic modification.

  #51052 commented on Mar 26, 2025 • 0 new comments

• [Refactor]Rename NCCL-related items to comm_backend

  #51061 commented on Mar 26, 2025 • 0 new comments

• [Docs] Update docs to reflect CPU requests/limits change in KubeRay v1.3

  #51072 commented on Mar 26, 2025 • 0 new comments

• [data] Make Dataset.name/set_name public

  #51076 commented on Mar 26, 2025 • 0 new comments

• [DONOTMERGE] POC for Ray+torch.distributed

  #51078 commented on Mar 26, 2025 • 0 new comments

• Fix the grammar of the OOM killer error messages

  #51081 commented on Mar 26, 2025 • 0 new comments

• [do not merge] Add Daft to the Ray ecosystem page

  #51133 commented on Mar 26, 2025 • 0 new comments

• Reproducing MacOS x86_64 Test Failure w/ Custom Numpy Serializer for ndarrays

  #51143 commented on Mar 26, 2025 • 0 new comments

• [core] Implement a universal printer

  #51151 commented on Mar 26, 2025 • 0 new comments

• [Do Not Merge] Update the Test Script to Debug test_network_failure_e2e Flaky Test

  #51153 commented on Mar 26, 2025 • 0 new comments

• Bump axios from 0.21.4 to 1.8.2 in /python/ray/dashboard/client

  #51162 commented on Mar 26, 2025 • 0 new comments

• Bump jinja2 from 3.1.3 to 3.1.6 in /release

  #51216 commented on Mar 26, 2025 • 0 new comments

• Bump keras from 2.15.0 to 3.9.0 in /python

  #51256 commented on Mar 26, 2025 • 0 new comments

• [Train V2] Fold `v2.LightGBMTrainer` API into the public trainer class as an alternate constructor

  #51265 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Add placement strategy to `EnvRunner` creation.

  #51267 commented on Mar 26, 2025 • 0 new comments

• Bump @babel/helpers from 7.19.4 to 7.26.10 in /python/ray/dashboard/client

  #51268 commented on Mar 26, 2025 • 0 new comments

• (WIP) [core][compiled graphs] Unify code paths for NCCL P2P and collectives scheduling

  #48649 commented on Mar 26, 2025 • 0 new comments

• [core][compiled graphs] Inter-execution overlap

  #48659 commented on Mar 26, 2025 • 0 new comments

• [Core]: Fix ConnectionError on Autoscaler CR lookups in K8s clusters …

  #48675 commented on Mar 26, 2025 • 0 new comments

• [Tune] Add OSS Vizier to Ray Tune

  #48684 commented on Mar 26, 2025 • 0 new comments

• [core] Fix building Ray against modern Protobuf versions

  #48724 commented on Mar 26, 2025 • 0 new comments

• [core] Enable STDERR custom formatting

  #48742 commented on Mar 26, 2025 • 0 new comments

• docs: update ray tune section

  #48769 commented on Mar 26, 2025 • 0 new comments

• docs: update ray serve section

  #48770 commented on Mar 26, 2025 • 0 new comments

• removed limit on log sizes via sockets

  #48780 commented on Mar 26, 2025 • 0 new comments

• [Fix][GCS] Implement reconnection for RedisContext

  #48781 commented on Mar 26, 2025 • 0 new comments

• [core][autoscaler]Reset the failure count to avoid RayCluster aborting unexpectedly

  #48797 commented on Mar 26, 2025 • 0 new comments

• [Data] Cleaned up & streamlined boundary sampling sequence to avoid conversion from Numpy to Python objects

  #48825 commented on Mar 26, 2025 • 0 new comments

• [Build][Deps] Add new `ray[azure]` extra package

  #48847 commented on Mar 26, 2025 • 0 new comments

• [core] cpp lint of object_manager

  #48878 commented on Mar 26, 2025 • 0 new comments

• [Autoscaler][Placement Group] Skip placed bundle when requesting resource

  #48924 commented on Mar 26, 2025 • 0 new comments

• [train] Make dataset argument covariant

  #48999 commented on Mar 26, 2025 • 0 new comments

• [core] Lint cpp files in common

  #49002 commented on Mar 26, 2025 • 0 new comments

• Slo track

  #49007 commented on Mar 26, 2025 • 0 new comments

• support to clean worker table with maximum_gcs_dead_worker_cached_count

  #49030 commented on Mar 26, 2025 • 0 new comments

• [Jobs] Add metric to track duration of jobs

  #49035 commented on Mar 26, 2025 • 0 new comments

• [WIP][compiled graphs] Avoid extra data I/O if CPU data is static

  #49042 commented on Mar 26, 2025 • 0 new comments

• [core][compiled-graphs] Very hacky Gloo channel PoC

  #49103 commented on Mar 26, 2025 • 0 new comments

• [Core] Persist the Driver Console Log When Job Execution Not Through Job API

  #49452 commented on Mar 26, 2025 • 0 new comments

• :bug: do not modify user-provided runtime_env

  #48021 commented on Mar 26, 2025 • 0 new comments

• [RLlib; Offline RL] - Enable gpu inference on data workers.

  #48041 commented on Mar 26, 2025 • 0 new comments

• [WIP][core] C++20 upgrade

  #48044 commented on Mar 26, 2025 • 0 new comments

• [core] Add metrics for Task RSS HWM.

  #48052 commented on Mar 26, 2025 • 0 new comments

• [gRPC] Fixing gRPC Server Call to be instantiated immediately for unbounded handlers

  #48057 commented on Mar 26, 2025 • 0 new comments

• [data] preprocessor: use map_batches in MaxAbsScaler, MinMaxScaler, UniformKBinsDiscretizer

  #48097 commented on Mar 26, 2025 • 0 new comments

• [Data] Fix a test that checks the "eliminate_build_output_blocks" optimization

  #48119 commented on Mar 26, 2025 • 0 new comments

• [Data] Fix a bug in the ReorderRandomizeBlocksRule optimization rule

  #48258 commented on Mar 26, 2025 • 0 new comments

• Add kuberay operator addon to cmd in gke-gcs-bucket.md

  #48268 commented on Mar 26, 2025 • 0 new comments

• [doc] Remove unused/unmaintained `doc/source/templates` folder

  #48295 commented on Mar 26, 2025 • 0 new comments

• [doc] fix: Typo and missing import in doc

  #48311 commented on Mar 26, 2025 • 0 new comments

• Fix invalid type for progress_reporter parameter of RunConfig

  #48439 commented on Mar 26, 2025 • 0 new comments

• [Data] Fix block accessors' combine handling of duplicate columns

  #48495 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix remote env runner request spam

  #48499 commented on Mar 26, 2025 • 0 new comments

• [runtime env]: Integrating Omnitrace to Ray worker process

  #48525 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix Algorithm with tune

  #48529 commented on Mar 26, 2025 • 0 new comments

• Add Adaptive Scaling Feature for Distributed Task Scheduling

  #48537 commented on Mar 26, 2025 • 0 new comments

• Overlap dynamic

  #48545 commented on Mar 26, 2025 • 0 new comments

• [data] add opensearch datasource

  #48555 commented on Mar 26, 2025 • 0 new comments

• [Docs][Collective] Fix examples to use init_collective_group and create_collective_group

  #48570 commented on Mar 26, 2025 • 0 new comments

• [core] Introduces Postable for InternalKVInterface.

  #48584 commented on Mar 26, 2025 • 0 new comments

• Dag bind order execution fix

  #48603 commented on Mar 26, 2025 • 0 new comments

• [Core][Compiled Graph] Execute DAG on Actor's Main Thread

  #48608 commented on Mar 26, 2025 • 0 new comments

• [Tune] Fix pbt restore in synch mode

  #48616 commented on Mar 26, 2025 • 0 new comments

• [core][cgraph] Use threadpool and one io_context for mutable object provider

  #49500 commented on Mar 26, 2025 • 0 new comments

• [train] add test for ScalingConfigV2 import

  #49515 commented on Mar 26, 2025 • 0 new comments

• [ci] Remove redundant ML doctests from running in unit test pipelines

  #49516 commented on Mar 26, 2025 • 0 new comments

• Overlap check deps

  #49520 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Add NPU and HPU support to RLlib

  #49535 commented on Mar 26, 2025 • 0 new comments

• [core][cgraph] Use cv instead of busy wait for next version

  #49542 commented on Mar 26, 2025 • 0 new comments

• [core] Minor improvements to core worker get

  #49567 commented on Mar 26, 2025 • 0 new comments

• [core] Don't get dashboard address after each dashboard connection failure

  #49584 commented on Mar 26, 2025 • 0 new comments

• [Core] Streaming generator supports num_returns

  #49586 commented on Mar 26, 2025 • 0 new comments

• [RLlib; docs] Docs do-over (new API stack): New `debugging.rst` page.

  #49592 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Support multiple accelerator monitoring and flexible display

  #49610 commented on Mar 26, 2025 • 0 new comments

• [core][docs] Lint some top level core docs

  #49703 commented on Mar 26, 2025 • 0 new comments

• [Core] Add virtual cluster

  #49717 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix broken stats accumulation for 'MeanStdFilter' connector.

  #49718 commented on Mar 26, 2025 • 0 new comments

• Update dyn-req-batch.md with style edits

  #49725 commented on Mar 26, 2025 • 0 new comments

• changes to get ray serve responding on REST API calls when distribute…

  #49730 commented on Mar 27, 2025 • 0 new comments

• [DATA]Add custom resources in data autoscaling

  #49756 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Flatten dict-typed observations before comparing them.

  #49758 commented on Mar 26, 2025 • 0 new comments

• [KubeRay] support suspending worker groups in KubeRay autoscaler

  #49768 commented on Mar 26, 2025 • 0 new comments

• [ci] remove pins in runtime_env usage in train examples

  #49772 commented on Mar 26, 2025 • 0 new comments

• Pass checkpointable args through in tf_learner

  #49861 commented on Mar 26, 2025 • 0 new comments

• New vsphere provider supporting Supervisor (k8s) cluster.

  #49881 commented on Mar 26, 2025 • 0 new comments

• [autoscaler] Fix potential dead lock in local provider

  #49909 commented on Mar 26, 2025 • 0 new comments

• Update azure.md - Missing azure dependency

  #49104 commented on Mar 26, 2025 • 0 new comments

• Fix memory issues caused by pyarrow.Dataset.to_batches.

  #49124 commented on Mar 26, 2025 • 0 new comments

• [compiled grapn][doc] structure

  #49134 commented on Mar 26, 2025 • 0 new comments

• [core] change all dynamic_pointer_cast to static_pointer_cast.

  #49135 commented on Mar 26, 2025 • 0 new comments

• [core] Gcs asio minor improvements

  #49169 commented on Mar 26, 2025 • 0 new comments

• [core][compiled-graphs] Gloo group

  #49187 commented on Mar 26, 2025 • 0 new comments

• [data] fix nodeName When the network in KubeRay is set to hostnetwork

  #49188 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] stop ray submmited job through ui

  #49201 commented on Mar 26, 2025 • 0 new comments

• Fix unpacking zip package treats "../" as the top_level_directory

  #49204 commented on Mar 26, 2025 • 0 new comments

• [WIP] Remove VM cluster autoscaler docker implementation

  #49238 commented on Mar 26, 2025 • 0 new comments

• [data] feat: Implement `ray.data.Dataset.offset`

  #49274 commented on Mar 26, 2025 • 0 new comments

• [Data] Fix numpy to arrow conversion.

  #49293 commented on Mar 27, 2025 • 0 new comments

• [wandb] Use wandb Run as a context manager

  #49307 commented on Mar 26, 2025 • 0 new comments

• [Core] fail to download s3 py modules

  #49332 commented on Mar 26, 2025 • 0 new comments

• [Serve] Improve serve deploy ignore behavior

  #49336 commented on Mar 26, 2025 • 0 new comments

• [Fix][Core] Periodically check log message queue cleared before shutdown

  #49337 commented on Mar 26, 2025 • 0 new comments

• [wip] Moving around

  #49345 commented on Mar 26, 2025 • 0 new comments

• Update tune-search-spaces.rst to correct outdated api use

  #49386 commented on Mar 26, 2025 • 0 new comments

• Adding input validation to ScalingConfig resources_per_worker

  #49389 commented on Mar 26, 2025 • 0 new comments

• [Draft] [spark] Set "HOST_IP" environmental variable for Ray worker nodes

  #49403 commented on Mar 26, 2025 • 0 new comments

• [core][compiled graphs] Support reduce scatter and all gather collective in compiled graph

  #49404 commented on Mar 26, 2025 • 0 new comments

• [core][dashboard] Dashboard head modules as Actors.

  #49432 commented on Mar 26, 2025 • 0 new comments

• [core][compiled-graphs] CachedChannel's inner channel must be provided

  #49434 commented on Mar 26, 2025 • 0 new comments

• [data] fix random_sample return different data in fixed seed

  #49443 commented on Mar 26, 2025 • 0 new comments

• [core] Improve Process management in Raylet

  #35252 commented on Mar 24, 2025 • 0 new comments

• [CI] `windows://python/ray/serve:test_metrics` is failing/flaky on master.

  #35452 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Windows CLI, cmd.exe, powershell parsing json arguments JSONDecodeError

  #35492 commented on Mar 24, 2025 • 0 new comments

• [Tune] Better force UTF8 encoding when calling Open method

  #34679 commented on Mar 24, 2025 • 0 new comments

• Ray can't use my resources correctly for parallelizing with OptunaSearch/Pandas/NumPy

  #34834 commented on Mar 24, 2025 • 0 new comments

• Build fails on ppc64le architecture

  #4309 commented on Mar 24, 2025 • 0 new comments

• [autoscaler] "Cannot perform an interactive login from a non TTY device" when trying to use a private docker registry

  #7339 commented on Mar 24, 2025 • 0 new comments

• Invalid memory access in RedisAsioClient/RedisAsyncContext on shutdown

  #9074 commented on Mar 24, 2025 • 0 new comments

• ray::IDLE processes persist if I disconnect and kill master process from IDE

  #9528 commented on Mar 24, 2025 • 0 new comments

• Unable to connect to ray head running on linux from ray worker node on windows

  #10362 commented on Mar 24, 2025 • 0 new comments

• Windows debugging on gdb does not work

  #9827 commented on Mar 24, 2025 • 0 new comments

• [RFC] Logging shutdown process to all Ray components.

  #13241 commented on Mar 24, 2025 • 0 new comments

• [CI] Upload Windows Status to flakey-tests.ray.io

  #12168 commented on Mar 24, 2025 • 0 new comments

• Cannot call remote instance method of a superclass from within a different instance method of the superclass

  #10899 commented on Mar 24, 2025 • 0 new comments

• [Bug] [Core] Unable to schedule fractional gpu jobs

  #20933 commented on Mar 24, 2025 • 0 new comments

• __del__ magic method can't access class properties

  #14285 commented on Mar 24, 2025 • 0 new comments

• [runtime env] raise exception for unsupported runtime_env features on Windows

  #21435 commented on Mar 24, 2025 • 0 new comments

• [Bug] [RLlib] Custom metrics are not reported to Tune

  #20938 commented on Mar 24, 2025 • 0 new comments

• [RLlib] League based PolicyMap across workers impacting scalability via memory use - Question/[Bug]

  #21459 commented on Mar 24, 2025 • 0 new comments

• [Bug] MultiDiscrete very slow

  #22507 commented on Mar 24, 2025 • 0 new comments

• Assertion Error on Seq lens for PPO with Attention only in evaluation.

  #22266 commented on Mar 24, 2025 • 0 new comments

• [RLlib] PPO - ray.rllib.agents.ppo "Put Error"

  #24307 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Metrics not reported with Client/Server and env=None

  #24601 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Dequeue check() returned False

  #25783 commented on Mar 24, 2025 • 0 new comments

• [core][gpu-objects] Support multiple tensors

  #51550 commented on Mar 25, 2025 • 0 new comments

• [Ray core] surface the node id/ip and other info (task name, etc.) in the stacktrace for a full object store

  #50408 commented on Mar 22, 2025 • 0 new comments

• Building an executable using Ray and Cx_freeze

  #42101 commented on Mar 24, 2025 • 0 new comments

• [dreamerv3] Get error when tuning custom env using dreamerv3

  #42107 commented on Mar 24, 2025 • 0 new comments

• SAC Checkpoint Loading Error

  #42651 commented on Mar 24, 2025 • 0 new comments

• [RLLib] custom TorchRLModule return action_dist but this results in an error

  #42786 commented on Mar 24, 2025 • 0 new comments

• [rllib] How to evaluate rollouts when using frame stacking RLModule?

  #42931 commented on Mar 24, 2025 • 0 new comments

• Core: ray.remote raises ValueError when used on torch IterableDataset

  #42914 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Calling AlgorithmConfig.build() for different algorithms inside the same execution context causes hard to debug issues.

  #43087 commented on Mar 24, 2025 • 0 new comments

• Ray with PyInstaller

  #27421 commented on Mar 24, 2025 • 0 new comments

• [Tune][Air] MLFlow Callback is incompatible with PB2

  #27783 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Resuming from checkpoint with DQN and epsilon greedy let timesteps start from 0 again

  #28289 commented on Mar 24, 2025 • 0 new comments

• [RLlib] The “trajectory_view_api” does not support the DQN algorithm, and the program will run in error

  #27609 commented on Mar 24, 2025 • 0 new comments

• [Jobs] Run jobs tests on Windows

  #28316 commented on Mar 24, 2025 • 0 new comments

• [CI] A simple way to reproduce osx/linux/windows CI run failure locally

  #29068 commented on Mar 24, 2025 • 0 new comments

• [Core] inspect_serializability bug - parent object serializable but bound method not

  #29423 commented on Mar 24, 2025 • 0 new comments

• [Core] util.multiprocessing.pool scheduling inefficiencies, blocking behavior in imap and imap_unordered

  #29453 commented on Mar 24, 2025 • 0 new comments

• [Core] Reference leakage somewhere after ray.shutdown()

  #30089 commented on Mar 24, 2025 • 0 new comments

• [Core] util.multiprocessing.pool: imap and imap_unordered blocking on ray.wait even though processes are complete

  #29466 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Undesired memory growing when using convolutional neural network

  #29699 commented on Mar 24, 2025 • 0 new comments

• [Core] Access violation on windows 11 when running modin workload

  #30493 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Using (gym) discrete and box spaces inside dict observation space throws ValueError: Expected flattened obs shape ...

  #31525 commented on Mar 24, 2025 • 0 new comments

• Windows python can not open file default_worker.py path with space

  #33047 commented on Mar 24, 2025 • 0 new comments

• [Core] RFC: simplify CI testing

  #34315 commented on Mar 24, 2025 • 0 new comments

• [Core] Incorrect detection of cpus

  #34846 commented on Mar 24, 2025 • 0 new comments

• [RLlib] `ExternalMultiAgentEnv` yields error on `log_returns` with `multiagent_done_dict`

  #35189 commented on Mar 24, 2025 • 0 new comments

• [core][gpu-objects] Object contains multiple tensors and/or mix of CPU data and GPU tensors

  #51274 commented on Mar 25, 2025 • 0 new comments

• [core][gpu-objects] Garbage collection for in-actor GPU objects

  #51262 commented on Mar 25, 2025 • 0 new comments

• [Core] Deserialization of generic pydantic models

  #47840 commented on Mar 25, 2025 • 0 new comments

• [Core] RayCheck failed: placement_group_resource_manager_->ReturnBundle(bundle_spec) Status not OK

  #51124 commented on Mar 25, 2025 • 0 new comments

• [Autoscaler][V2] Updating max replicas while Pods are pending causes v2 autoscaler to hang

  #50868 commented on Mar 25, 2025 • 0 new comments

• [data] importing ray.data closes logging handlers, breaking custom logging

  #48846 commented on Mar 26, 2025 • 0 new comments

• [Core|Dataset] Ray job stuck with idle actors with no tasks

  #45822 commented on Mar 26, 2025 • 0 new comments

• [Core] Default concurrency using concurrency groups

  #46666 commented on Mar 26, 2025 • 0 new comments

• [Data] Adding streaming capability for `ray.data.Dataset.unique`

  #51207 commented on Mar 26, 2025 • 0 new comments

• [Data] Filter operation changes schema of dataset

  #51217 commented on Mar 26, 2025 • 0 new comments

• Core: Ray cluster nodes underutilization during autoscaling

  #47355 commented on Mar 26, 2025 • 0 new comments

• [RFC] [Serve] Custom Scaling

  #41135 commented on Mar 26, 2025 • 0 new comments

• [Feedback] Feedback for ray + uv

  #50961 commented on Mar 26, 2025 • 0 new comments

• CI test linux://rllib:learning_tests_cartpole_dqn_gpu is flaky

  #46683 commented on Mar 26, 2025 • 0 new comments

• [Serve] Consider custom resources in best-fit node selection for DeploymentScheduler in Ray Serve

  #51361 commented on Mar 26, 2025 • 0 new comments

• [Core] Python 3.13 wheel

  #49738 commented on Mar 26, 2025 • 0 new comments

• [train v2][tune] Migration Guide

  #49454 commented on Mar 26, 2025 • 0 new comments

• CI test linux://rllib:learning_tests_multi_agent_pendulum_sac_multi_gpu is flaky

  #47309 commented on Mar 26, 2025 • 0 new comments

• [Data] - read_parquet raises AWS Error NETWORK_CONNECTION during HeadObject operation: curlCode: 43

  #35826 commented on Mar 26, 2025 • 0 new comments

• [Data] Progress bars are sometimes half completed even after the task is finished

  #36490 commented on Mar 26, 2025 • 0 new comments

• [Data] Progress Bars are incomplete (notebooks + terminal)

  #36181 commented on Mar 26, 2025 • 0 new comments

• [Data] Ray 2.6 created a breaking change in the index of a Modin DataFrame

  #37771 commented on Mar 26, 2025 • 0 new comments

• [data] ray_tqdm does not work with numba

  #45538 commented on Mar 26, 2025 • 0 new comments

• [Data] Row outputted returns 0

  #48484 commented on Mar 26, 2025 • 0 new comments

• Ray component: Core: PoolActor processes hanging

  #24784 commented on Mar 24, 2025 • 0 new comments

• Ray on kubernetes with custom image_uri is broken

  #51423 commented on Mar 24, 2025 • 0 new comments

• [Autoscaler, data] Ray starts `AutoscalingRequester` even when using `enableInTreeAutoscaling`

  #51559 commented on Mar 24, 2025 • 0 new comments

• [Core] build from source code guide is out of date

  #43093 commented on Mar 25, 2025 • 0 new comments

• [cgraph] Support ray.wait() for CompiledDAGRef

  #51391 commented on Mar 25, 2025 • 0 new comments

• [Workflow] Add Azure as one of the storage backend options?

  #34910 commented on Mar 25, 2025 • 0 new comments

• [Train] Intermittent `UnpicklingError` when loading estimator/preprocessor from checkpoint

  #33815 commented on Mar 25, 2025 • 0 new comments

• [Core] std::bad_alloc error using ray.init()

  #33525 commented on Mar 25, 2025 • 0 new comments

• [RLlib] APPO gets extremely slow when run with >1 GPUs.

  #50221 commented on Mar 25, 2025 • 0 new comments

• [RLlib] Tuner.restore() Not Restoring Training

  #43266 commented on Mar 25, 2025 • 0 new comments

• [BUG] Ray dashboard client failed to build

  #23548 commented on Mar 25, 2025 • 0 new comments

• [Core][Streaming generator] Support num_returns.

  #46934 commented on Mar 25, 2025 • 0 new comments

• [llm] Roadmap for Data and Serve LLM APIs

  #51313 commented on Mar 25, 2025 • 0 new comments

• [Serve] Detailed Analysis of Errors Related to 'Ray does not allocate any GPUs on the driver node' && 'No CUDA GPUs are available'

  #51242 commented on Mar 25, 2025 • 0 new comments

• [Ray dashboard] Actors tab does not list actors under certain conditions

  #47447 commented on Mar 25, 2025 • 0 new comments

• [<Ray component: Core|RLlib|etc...>] Support for Gradio version 4 on Ray Serve

  #49245 commented on Mar 25, 2025 • 0 new comments

• [RFC] Async request support in Ray Serve

  #32292 commented on Mar 25, 2025 • 0 new comments

• [Serve] Observability for proxy

  #48184 commented on Mar 25, 2025 • 0 new comments

• [Serve] ingress decorator does not work with fastapi.APIRouter arg

  #50372 commented on Mar 25, 2025 • 0 new comments

• [Serve] FastAPI ingress does not work with composable routers

  #50373 commented on Mar 25, 2025 • 0 new comments

• [core][gpu-objects] IPC communication for processes on the same GPU

  #51270 commented on Mar 25, 2025 • 0 new comments

• [core][gpu-objects] CollectiveGroupManager

  #51260 commented on Mar 25, 2025 • 0 new comments

• [core][gpu-objects] CollectiveExecutor

  #51261 commented on Mar 25, 2025 • 0 new comments

• [core][gpu-objects] Driver should order all collective calls to avoid deadlock

  #51264 commented on Mar 25, 2025 • 0 new comments

• [RLlib] Silence external warnings

  #24107 commented on Mar 25, 2025 • 0 new comments

• [Core] Spot preemption related retries do not count towards the max retries

  #50640 commented on Mar 22, 2025 • 0 new comments

• [Core] Plugable storage backend besides Redis

  #50656 commented on Mar 22, 2025 • 0 new comments

• [nsys plugin] How about add an option `name` to nsys dumped file

  #50711 commented on Mar 22, 2025 • 0 new comments

• [Ray Core] Slow scheduling speed with IOError: Broken pipe

  #50244 commented on Mar 22, 2025 • 0 new comments

• [Core] ray.util.ActorPool can get stuck in failing state with one bad actor

  #50313 commented on Mar 22, 2025 • 0 new comments

• [compiled graph] Driver cannot participate in the NCCL group

  #50423 commented on Mar 22, 2025 • 0 new comments

• [Core] Negative available resources

  #50739 commented on Mar 22, 2025 • 0 new comments

• [core] Serve microbenchmarks occasionally crash with segfault or invalid memory access

  #50802 commented on Mar 22, 2025 • 0 new comments

• [Core] Ray Data job hanging with flooded Cancelling stale RPC with seqno 125 < 127 error

  #50814 commented on Mar 22, 2025 • 0 new comments

• [Core] calling remote function in `Future` callback breaks ray

  #50980 commented on Mar 22, 2025 • 0 new comments

• [core] question about ray issue: 51051

  #51554 commented on Mar 22, 2025 • 0 new comments

• [Ray component: Python|runtime_env]Pip install `whl` file faliure when a job reruns in the same cluster

  #49059 commented on Mar 22, 2025 • 0 new comments

• [Data]: Categorizer fails with non uniform distributions

  #50792 commented on Mar 22, 2025 • 0 new comments

• [core] ray.init does not work with local_mode on run_time envs.

  #30273 commented on Mar 23, 2025 • 0 new comments

• [RLlib] Basic PPO script throws obscure error when building RLModule

  #51333 commented on Mar 23, 2025 • 0 new comments

• [core] Split giant ray core C++ targets into small ones

  #50586 commented on Mar 24, 2025 • 0 new comments

• [Train] Deepspeed + Triton 3.2.0 + Torch 2.6.0 has issues with Ray Train

  #50406 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Attribute error when trying to compute action after training Multi Agent PPO with New API Stack

  #44475 commented on Mar 24, 2025 • 0 new comments

• [AIR] Sampling support for Ray Train/Ray Data

  #31127 commented on Mar 24, 2025 • 0 new comments

• [Serve/Core] Raylet crash encountered in Serve during Actor termination

  #51408 commented on Mar 24, 2025 • 0 new comments

• [Ray Data | Core ]

  #51416 commented on Mar 24, 2025 • 0 new comments

• Cannot Install ray[rllib] on Python 3.13

  #50226 commented on Mar 24, 2025 • 0 new comments

• for exporting r2d2+lstm to onnx, why is empty state being passed in?

  #50166 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Can't create multi-agent external env with new API stack

  #46961 commented on Mar 24, 2025 • 0 new comments

• [core] Ray session conflicts with PyArrow+HDFS

  #36415 commented on Mar 21, 2025 • 0 new comments

• [Core] Limited interoperability with JAX

  #46760 commented on Mar 21, 2025 • 0 new comments

• [Core] Too many threads in ray worker

  #36936 commented on Mar 21, 2025 • 0 new comments

• [Runtime Environment] Remove cached python libs, working dir etc

  #47488 commented on Mar 21, 2025 • 0 new comments

• [Data] Infinite recursion in ansitowin32.py (under tqdm_ray)

  #51337 commented on Mar 21, 2025 • 0 new comments

• [Data] `read_images` benchmark sometimes fails with `ArrowVariableShapedTensorArray` error

  #49883 commented on Mar 21, 2025 • 0 new comments

• [Core] ray raises a "Failed to unpickle serialized exception" error when an OpenAI Authentication Error is raised in task

  #43428 commented on Mar 21, 2025 • 0 new comments

• [core] Set OPENBLAS_NUM_THREADS to number of cpus automatically

  #34724 commented on Mar 21, 2025 • 0 new comments

• [Core] Getting node id for usage in NodeAffinitySchedulingStrategy

  #28195 commented on Mar 21, 2025 • 0 new comments

• [usability][Feature] Throw error message if resolved ip address doesn't match the localhost

  #19052 commented on Mar 21, 2025 • 0 new comments

• [StateAPI] StateAPI request truncates recent elements

  #50378 commented on Mar 21, 2025 • 0 new comments

• [core] ray.remote Decorator's Return Type Cannot Be Determined by Type Checkers

  #50410 commented on Mar 21, 2025 • 0 new comments

• [Feature] [Performance] [Docs] Disabling object spilling is not documented

  #21998 commented on Mar 21, 2025 • 0 new comments

• [Core] classmethod support for actors

  #36986 commented on Mar 21, 2025 • 0 new comments

• [data] Cannot convert dict to PyArrow blocks

  #42075 commented on Mar 21, 2025 • 0 new comments

• [Core] `ray.cancel` multiple ObjectRefs

  #24559 commented on Mar 21, 2025 • 0 new comments

• [core][cluster launcher] Cluster launcher should use `docker run --gpus` if GPUs are autodetected on the worker node

  #43231 commented on Mar 21, 2025 • 0 new comments

• [core][cluster launcher] Local clusters should stop Ray containers on `ray down`

  #43232 commented on Mar 21, 2025 • 0 new comments

• [Core] Enable huge pages for object store

  #51352 commented on Mar 21, 2025 • 0 new comments

• [Dashboard] Fix listing APIs to avoid truncating at 10k entities

  #48251 commented on Mar 21, 2025 • 0 new comments

• [core] Upgrade grpc (Mar.15, 2025)

  #51395 commented on Mar 21, 2025 • 0 new comments

• [core][gpu-objects] Driver tries to get the data from in-actor store

  #51272 commented on Mar 21, 2025 • 0 new comments

• [RLlib] Possible bug in TorchSquashedGaussian plus associated feature request

  #51544 commented on Mar 21, 2025 • 0 new comments

• [Core] get_user_temp_dir() Doesn't Honor the User Specified Temp Dir

  #51218 commented on Mar 21, 2025 • 0 new comments

• [Core] Please provide better message where 'RuntimeError: Failed to unpickle serialized exception'

  #49885 commented on Mar 21, 2025 • 0 new comments

• [Tune] Cannot run the QuickStart example code on windows after installing Ray in conda enviroment, reporting FileNotFoundError

  #46827 commented on Mar 24, 2025 • 0 new comments

• RLlib: dist_class is missed while I try to use Policy.learn_on_batch()

  #47011 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Flatten observations example doesn't work

  #47127 commented on Mar 24, 2025 • 0 new comments

• Issue after pip install of ray tune

  #47266 commented on Mar 24, 2025 • 0 new comments

• Install Ray version 1.5.2

  #46776 commented on Mar 24, 2025 • 0 new comments

• ray issue

  #47177 commented on Mar 24, 2025 • 0 new comments

• [RLLib] Expected scalars to be on CPU, got cuda:0 instead

  #35640 commented on Mar 24, 2025 • 0 new comments

• [RLlib] PPO Memory Leak on Uneven CNN (conv) filters

  #35866 commented on Mar 24, 2025 • 0 new comments

• [Rllib][Tune] AttributeError: 'TorchCategorical' object has no attribute 'log_prob' with PB2

  #35923 commented on Mar 24, 2025 • 0 new comments

• [Tune] PB2 Checkpoint/Sync Path Compatibility with Windows

  #36370 commented on Mar 24, 2025 • 0 new comments

• [Release process] Validating and uploading wheels is a pain and a error prone

  #36522 commented on Mar 24, 2025 • 0 new comments

• [<Ray component: cluster>] Urllib3 warning messages cannot be blocked in Ray

  #36577 commented on Mar 24, 2025 • 0 new comments

• [JAVA] Ray.init() failed when JNI load

  #36637 commented on Mar 24, 2025 • 0 new comments

• UnicodeEncodeError: 'charmap' codec can't encode characters in position 0-68: character maps to <undefined>

  #36767 commented on Mar 24, 2025 • 0 new comments

• Unable to build Ray on Power (Error: key "3.9.16" not found in dictionary)

  #37889 commented on Mar 24, 2025 • 0 new comments

• [RLlib|Tune] Cannot restore tune checkpoints in algorithm

  #39785 commented on Mar 24, 2025 • 0 new comments

• [Core] Ray is slower than serial python

  #40184 commented on Mar 24, 2025 • 0 new comments

• [RLlib] Torch Autoregressive Example does not work with gpu

  #38645 commented on Mar 24, 2025 • 0 new comments

• [Tune] Font color in dark theme Jupyter Notebooks in VS Code

  #40317 commented on Mar 24, 2025 • 0 new comments

• CheckpointConfig does not work on Windows

  #37226 commented on Mar 24, 2025 • 0 new comments

• [PPOConfig] Utilising new API/models without matching documentation

  #40201 commented on Mar 24, 2025 • 0 new comments

• [core] Recent windows test flakiness

  #38413 commented on Mar 24, 2025 • 0 new comments

• [tune] Full /tmp/ folder, ray does not clean up

  #41202 commented on Mar 24, 2025 • 0 new comments

• [Core] StreamingObjectRefGenerator not working over network

  #41556 commented on Mar 24, 2025 • 0 new comments

• Upgrade Windows CI docker image to use Windows 11 and more recent toolchains.

  #49830 commented on Mar 24, 2025 • 0 new comments

• Upgrade windows CI AMI to use Windows 11

  #49829 commented on Mar 24, 2025 • 0 new comments

• [RAY TRAIN] Force use of gloo in Windows

  #49778 commented on Mar 24, 2025 • 0 new comments

• [RLlib][Windows] Windows Invalid Directory Name Error in Ray RLlib

  #49477 commented on Mar 24, 2025 • 0 new comments

• [Data] Transient Parquet Fragment Serialization Error

  #49082 commented on Mar 24, 2025 • 0 new comments

• BUILD: patch zlib for macos and protobuf for windows

  #48794 commented on Mar 24, 2025 • 0 new comments

• [Distributed Debugger] Newly added breakpoint not works: Breakpoint in file that does not exist

  #48778 commented on Mar 24, 2025 • 0 new comments

• [Ray + YOLOv8] YOLOv8 model.tune

  #47859 commented on Mar 24, 2025 • 0 new comments

• [tune] Repeated runs don't get averaged by search algorithm

  #47758 commented on Mar 24, 2025 • 0 new comments

• |RLlib] New API Stack: "local_gpu_idx 0 is not a valid GPU id or is not available."

  #47364 commented on Mar 24, 2025 • 0 new comments

• [Core] Warning message output as error cannot be filtered/hidden; unexposed environmental variable

  #43264 commented on Mar 24, 2025 • 0 new comments

• [BUG] Ray crashes my python process when the connected kernel goes away.

  #43280 commented on Mar 24, 2025 • 0 new comments

• [RLlib + Tune] PermissionError: [WinError 5] Access is denied: '../.tmp_generator' -> '..basic-variant-state-..' while training with ``Tuner``

  #43702 commented on Mar 24, 2025 • 0 new comments

• [RLlib] step() function called too early, could lead to inconsistencies

  #44290 commented on Mar 24, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_actor_retry is flaky

  #43845 commented on Mar 24, 2025 • 0 new comments

• [<Ray component: syncer.py.>] Last sync command failed with the following error

  #44320 commented on Mar 24, 2025 • 0 new comments

• [RLlib] make_multi_callbacks with new API stack error

  #44386 commented on Mar 24, 2025 • 0 new comments

• [RLlib] EagerTFPolicyV2 wrongly calls overridden action_sampler_fn

  #44671 commented on Mar 24, 2025 • 0 new comments

• [RLlib] New API Stack: Action masking not working with wrapper, default encoder config issue

  #44780 commented on Mar 24, 2025 • 0 new comments

• Ray Weekly Release

  #44276 commented on Mar 24, 2025 • 0 new comments

• [RLlib] TypeError converting batch (INFOS) to torch tensor with ConnectorV2

  #44478 commented on Mar 24, 2025 • 0 new comments

• tmp directory path issue between Windows client and Linux Ray cluster head

  #45010 commented on Mar 24, 2025 • 0 new comments

• Ray tune on Mac M2/M1 never stop

  #45797 commented on Mar 24, 2025 • 0 new comments

• [<Ray component: Serve>] Worker node is killed after starting with reason of missing too many heartbeat checks

  #46548 commented on Mar 24, 2025 • 0 new comments

• [tune] Can't find driver_artifacts file

  #46607 commented on Mar 24, 2025 • 0 new comments

• [WIP][Feature commit] Initial commit for supporting IPv6 stack in Ray Clus…

  #40332 commented on Mar 27, 2025 • 0 new comments

• [core] Tidying up mmap and munmap a bit.

  #40334 commented on Mar 26, 2025 • 0 new comments

• [Ray Train] Implement strict even-split of training workers for pretraining

  #40442 commented on Mar 26, 2025 • 0 new comments

• Add dolly v2 instruction tuning ray train

  #40455 commented on Mar 26, 2025 • 0 new comments

• [Doc] Add note in `ray submit` doc to recommend Ray Job API

  #40500 commented on Mar 26, 2025 • 0 new comments

• [dashboard] ignore reinit error when getting dashboard url

  #40545 commented on Mar 26, 2025 • 0 new comments

• [tune] link placement group doc

  #40590 commented on Mar 26, 2025 • 0 new comments

• [Train] Support rank_zero_only uploading for Lightning RayTrainReportCallback

  #40639 commented on Mar 26, 2025 • 0 new comments

• [Core] Add observability support to AcceleratorManager

  #40749 commented on Mar 26, 2025 • 0 new comments

• [core] Fix windows conda activate with conda.bat as executable in conda path

  #40779 commented on Mar 26, 2025 • 0 new comments

• [Serve] Fix Windows unit tests

  #40812 commented on Mar 27, 2025 • 0 new comments

• [Core] [Cluster Launcher] Rename min/max_workers to min/max_worker_nodes

  #40835 commented on Mar 26, 2025 • 0 new comments

• [Doc] Clarify that a recent version of nsight is needed

  #40846 commented on Mar 26, 2025 • 0 new comments

• [Core] Add logs upon abrupt failure code path

  #40849 commented on Mar 26, 2025 • 0 new comments

• [docs] Documentation fixes (logging and profiling)

  #40915 commented on Mar 26, 2025 • 0 new comments

• [RFC v3] Ray Client2

  #40990 commented on Mar 26, 2025 • 0 new comments

• [WIP] accelerated DAG

  #40991 commented on Mar 26, 2025 • 0 new comments

• WIP Do Not Merge

  #41025 commented on Mar 26, 2025 • 0 new comments

• Adapt the joblib backend for compatibility with `return_as=generator`

  #41028 commented on Mar 26, 2025 • 0 new comments

• TPU pod autoscaling based on the TpuCommandRunner

  #41065 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Add SUPER algorithm.

  #41079 commented on Mar 26, 2025 • 0 new comments

• [docs][Serve] add text about pip-pack installation

  #41088 commented on Mar 26, 2025 • 0 new comments

• [Core] mig auto detection

  #41103 commented on Mar 26, 2025 • 0 new comments

• WIP recharts custom charting library

  #41140 commented on Mar 26, 2025 • 0 new comments

• Fix docker gpu 2

  #42426 commented on Mar 26, 2025 • 0 new comments

• Release test sort.regular failed

  #50417 commented on Mar 27, 2025 • 0 new comments

• [docs] try enabling nitpicky

  #39448 commented on Mar 26, 2025 • 0 new comments

• Fix for appo_torch_policy.py when used with attention_net

  #39520 commented on Mar 26, 2025 • 0 new comments

• [CherryPick][Serve] Ignore cancel request when receving websocket.accept message (#39413)

  #39625 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Issue 38560: New API (Learner) stack does not properly count steps trained/sampled.

  #39628 commented on Mar 26, 2025 • 0 new comments

• [DEBUG] metric

  #39659 commented on Mar 26, 2025 • 0 new comments

• Upgrade default AWS DLAMI

  #39721 commented on Mar 26, 2025 • 0 new comments

• [RLLib-contrib] Implementation of RED-Q (Ensemble SAC) Algorithm in PyTorch

  #39747 commented on Mar 26, 2025 • 0 new comments

• [core] Use futures to synchornize `parallel_memcopy`

  #39755 commented on Mar 26, 2025 • 0 new comments

• fix: add cython async detection

  #39762 commented on Mar 26, 2025 • 0 new comments

• [core][RFC] http based pure external client

  #39771 commented on Mar 26, 2025 • 0 new comments

• [Data] Consolidate default fault tolerance options

  #39797 commented on Mar 26, 2025 • 0 new comments

• [core] Fix a corner case where the RPC never return

  #39801 commented on Mar 26, 2025 • 0 new comments

• [Cluster launcher] Make cluster Ray version match client Ray version by default

  #39812 commented on Mar 26, 2025 • 0 new comments

• [core] Default actor object's callable method to ActorMethod.remote

  #39826 commented on Mar 26, 2025 • 0 new comments

• [Logging] Fix Deduplication URL

  #39830 commented on Mar 26, 2025 • 0 new comments

• [data] Fix map_batches on datasets with nested lists

  #39869 commented on Mar 26, 2025 • 0 new comments

• [CI] [Doc] Add reminder to install setup hooks for linter

  #39888 commented on Mar 26, 2025 • 0 new comments

• [Cluster launcher] disable verbose logs by default

  #39930 commented on Mar 26, 2025 • 0 new comments

• static DAGs

  #39956 commented on Mar 26, 2025 • 0 new comments

• [data] Make exceptions consistent when falling back to pandas

  #39969 commented on Mar 26, 2025 • 0 new comments

• [core][RFC v2] HTTP based Ray Client

  #40085 commented on Mar 26, 2025 • 0 new comments

• RFC: Add Julia Language support

  #40098 commented on Mar 26, 2025 • 0 new comments

• [train] update some imports from ray.air to ray.train

  #40171 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Adding action space scaling to Gaussian noise in exploration

  #40281 commented on Mar 26, 2025 • 0 new comments

• [WIP][Streaming] Cleaning up streaming sequence

  #42443 commented on Mar 26, 2025 • 0 new comments

• WIP [data] A streaming compatible implementation of repartition-by-column

  #42477 commented on Mar 26, 2025 • 0 new comments

• [tune] remove tensorboardX upper bound

  #42581 commented on Mar 26, 2025 • 0 new comments

• allow victoria metrics response message

  #42620 commented on Mar 26, 2025 • 0 new comments

• [WIP][Serve] Revisited cancellation handling in Proxy to make sure response generator is properly cancelled

  #42665 commented on Mar 26, 2025 • 0 new comments

• [DO NOT REVIEW, LONG TERM PR FOR CI] Pinterest main branch 2.9.1

  #42672 commented on Mar 26, 2025 • 0 new comments

• [WIP] Experimental batching support in Streaming Generators

  #42825 commented on Mar 26, 2025 • 0 new comments

• Fix None exception in evaluate.

  #42858 commented on Mar 26, 2025 • 0 new comments

• [Data] Block compression for ArrowBlock

  #42859 commented on Mar 27, 2025 • 0 new comments

• reduce lock mutex scope

  #43067 commented on Mar 26, 2025 • 0 new comments

• [Core/Accelerated DAG] Support Gloo-based backend using Ray collective group.

  #43096 commented on Mar 26, 2025 • 0 new comments

• [RLlib] ConnectorV2 API: Add heuristic action logits mixin example script.

  #43107 commented on Mar 26, 2025 • 0 new comments

• [Core] ray.remote raises ValueError when used on torch IterableDataset

  #43117 commented on Mar 26, 2025 • 0 new comments

• [docs] Add antipattern for nested ray.get

  #43184 commented on Mar 26, 2025 • 0 new comments

• [docs][clusters] Improve instructions for GPU autodetection and manual cluster launching

  #43219 commented on Mar 26, 2025 • 0 new comments

• Release performance regression 2.9.2/2.9.3

  #43235 commented on Mar 26, 2025 • 0 new comments

• [Build] Add build for RH

  #43335 commented on Mar 26, 2025 • 0 new comments

• [WIP] Use 3.9 in macos.

  #43351 commented on Mar 26, 2025 • 0 new comments

• [WIP] prepend pickled function co_fileanme with "<ray remote>"

  #43359 commented on Mar 26, 2025 • 0 new comments

• verify windows wheels.

  #43442 commented on Mar 26, 2025 • 0 new comments

• [Doc] Add RAY_REDIS_CA_CERT description for GCS fault tolerance

  #43478 commented on Mar 26, 2025 • 0 new comments

• [Core][CLI] fix ray status long decimal numbers

  #43480 commented on Mar 26, 2025 • 0 new comments

• [core] Fix max_calls option when used on a worker that is part of a workflow

  #43700 commented on Mar 26, 2025 • 0 new comments

• [WIP] Core worker shutdown then disconnect

  #43759 commented on Mar 26, 2025 • 0 new comments

• [Core] gpu memory scheduling prototype

  #41147 commented on Mar 26, 2025 • 0 new comments

• [RFC] Ray Client2 with original Ray APIs

  #41323 commented on Mar 26, 2025 • 0 new comments

• [core] Refactors the delayed task resubmission.

  #41351 commented on Mar 26, 2025 • 0 new comments

• Relax check_version_info to check for bytecode compatibility

  #41373 commented on Mar 26, 2025 • 0 new comments

• Update metrics.py

  #41385 commented on Mar 27, 2025 • 0 new comments

• [Core] Return the correct task ID when get_runtime_context is used in a background thread

  #41397 commented on Mar 26, 2025 • 0 new comments

• [serve] Adjust the doc of the Serve Java API

  #41398 commented on Mar 26, 2025 • 0 new comments

• [core] Vendor aiohttp and aiosignal for Ray.

  #41426 commented on Mar 26, 2025 • 0 new comments

• [train] update XGBoost model format to UBJ

  #41442 commented on Mar 26, 2025 • 0 new comments

• Feat/metric validation

  #41478 commented on Mar 26, 2025 • 0 new comments

• [Cluster Launcher] Update head node commands to refer to which node they can be run from

  #41490 commented on Mar 26, 2025 • 0 new comments

• docs: add user guide on KubeRay webhooks

  #41527 commented on Mar 26, 2025 • 0 new comments

• [Core] Track resource per instance [1/n]

  #41582 commented on Mar 26, 2025 • 0 new comments

• [doc] Add documentation guide for MPI on Ray.

  #41626 commented on Mar 26, 2025 • 0 new comments

• [RFC v4] Ray Client2.

  #41803 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Should specify the time range in job detail page for load the cluster status and scale metrics

  #41828 commented on Mar 26, 2025 • 0 new comments

• add disk throughput test

  #41882 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Significant performance improvement with curriculum learning when using a high number of rollout workers

  #41910 commented on Mar 26, 2025 • 0 new comments

• [WIP] Multinode dag

  #42059 commented on Mar 26, 2025 • 0 new comments

• Fix a bug where start_time could be None leading to a crash in TuneTerminalReporter

  #42078 commented on Mar 26, 2025 • 0 new comments

• [draft] 100tb shuffle experimental

  #42086 commented on Mar 26, 2025 • 0 new comments

• [WIP] Multinode DAG minus FT changes

  #42173 commented on Mar 26, 2025 • 0 new comments

• [Tune][Air] Fix MLflowLoggerCallback to enable its use with PBT (#27783)

  #42182 commented on Mar 26, 2025 • 0 new comments

• [rllib_contrib] Lagrangian PPO

  #42365 commented on Mar 26, 2025 • 0 new comments

• [WIP][Tracing] Fixing tracing context injection

  #42384 commented on Mar 26, 2025 • 0 new comments

• Release test sort.chaos failed

  #49765 commented on Mar 27, 2025 • 0 new comments

• Release test random_shuffle.chaos failed

  #49395 commented on Mar 27, 2025 • 0 new comments

• Release test random_shuffle.regular failed

  #49383 commented on Mar 27, 2025 • 0 new comments

• [Train] Crash at end of training

  #51527 commented on Mar 27, 2025 • 0 new comments

• [<Ray component: Core>] num_gpus not working with ROCM devices

  #46563 commented on Mar 27, 2025 • 0 new comments

• [Data] ray.data.from_torch fails on datasets with variable shaped images

  #50229 commented on Mar 27, 2025 • 0 new comments

• CI test linux://python/ray/tune:test_train_v2_integration is flaky

  #49930 commented on Mar 27, 2025 • 0 new comments

• [Core] Ray Label Selector API Implementation Tracker

  #51564 commented on Mar 27, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_actor_retry2 is flaky

  #47415 commented on Mar 27, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_reference_counting_2 is flaky

  #45964 commented on Mar 27, 2025 • 0 new comments

• [train] add model (pipeline) parallelism example

  #22894 commented on Mar 26, 2025 • 0 new comments

• [Air] add Jax trainer

  #25385 commented on Mar 26, 2025 • 0 new comments

• [RLlib] Fix Issue #25316: unconfigurable `dist_dim` for custom multi-action distributions

  #25490 commented on Mar 26, 2025 • 0 new comments

• [Core] Allow task retry for `ray.cancel`

  #26254 commented on Mar 26, 2025 • 0 new comments

• [RLLib][Air] MLFlow parsing of RLLib evaluation and custom metrics

  #26711 commented on Mar 26, 2025 • 0 new comments

• [State API] Add input & output size to the task API

  #31898 commented on Mar 26, 2025 • 0 new comments

• [core][bugfix] catch BaseException

  #32105 commented on Mar 26, 2025 • 0 new comments

• [core] Add support for one log per worker pool worker

  #33167 commented on Mar 26, 2025 • 0 new comments

• [Core] Allow user to specify DASHBOARD_AGENT_LISTEN_PORT

  #34886 commented on Mar 26, 2025 • 0 new comments

• Refactor run_release_test.sh

  #35065 commented on Mar 26, 2025 • 0 new comments

• [Serve] Increase the GCS timeout

  #35330 commented on Mar 26, 2025 • 0 new comments

• UX: rework stdout error when ray fails to start

  #35378 commented on Mar 26, 2025 • 0 new comments

• [Dashboard] Provide a job dashboard URL link instead of the dashboard link when ray.init is called.

  #35427 commented on Mar 26, 2025 • 0 new comments

• [1/2] [UI] Event Observability, add a new event table

  #38638 commented on Mar 27, 2025 • 0 new comments

• [Data] A proper way to handle random seed in random_sample() and other sampling for reproducibility

  #50638 commented on Mar 26, 2025 • 0 new comments

• [Ray Data] Introduce "Key concepts" to Ray Data doc

  #50018 commented on Mar 26, 2025 • 0 new comments

• [Data] Refactor `ParquetDatasink._write_partition_files` to use `pyarrow.parquet.write_to_dataset`

  #50502 commented on Mar 26, 2025 • 0 new comments

• [Data]Extend Ray Data with read/write hive

  #51094 commented on Mar 26, 2025 • 0 new comments

• [Data] supper passing `pyarrow.dataset.Expression`s to `Dataset.filter`'s `expr`

  #50799 commented on Mar 26, 2025 • 0 new comments

• [data] RefBundle doesn't always eagerly free data

  #37910 commented on Mar 26, 2025 • 0 new comments

• [Data] Add support for all Spark RDD transformations and actions

  #10983 commented on Mar 26, 2025 • 0 new comments

• [Data, Train] ray::SplitCoordinator is very slow at every epoch + takes up too much memory

  #49190 commented on Mar 26, 2025 • 0 new comments

• [Ray Data] Support S3 Tables

  #49083 commented on Mar 26, 2025 • 0 new comments

• CI test linux://rllib:examples/connectors/mean_std_filtering_ppo is flaky

  #47435 commented on Mar 26, 2025 • 0 new comments

• [Ray debugger] Unable to use debugger on slurm cluster

  #51157 commented on Mar 27, 2025 • 0 new comments

• [Core] Runtime env working_dir validation

  #51380 commented on Mar 27, 2025 • 0 new comments

• [<Ray component: java>] expose ObjectRef in DeploymentResponse class

  #51445 commented on Mar 27, 2025 • 0 new comments

• [Ray Serve]: "RuntimeError: No CUDA GPUs are available" when running vllm with ray

  #51193 commented on Mar 27, 2025 • 0 new comments

• [Serve] On kuberay, vLLM-0.7.2 reports "No CUDA GPUs are available" while vllm-0.6.6.post1 works fine when deploy rayservice

  #51154 commented on Mar 27, 2025 • 0 new comments

• [Ray Serve] Expose public interface for user to customize the router

  #50465 commented on Mar 27, 2025 • 0 new comments

• [Serve] Proxy actor not started on worker node when using kuberay

  #50349 commented on Mar 27, 2025 • 0 new comments

• [Serve] make various default values of `AutoscalingConfig.max_replicas` consistent and >1

  #50222 commented on Mar 27, 2025 • 0 new comments

• [Serve] exceptions raised by request timeout are inconsistent

  #50992 commented on Mar 27, 2025 • 0 new comments

• CI test darwin://python/ray/tests:test_threaded_actor is flaky

  #44663 commented on Mar 27, 2025 • 0 new comments

• [Core/Data] Name GPU Worker Processes

  #40529 commented on Mar 27, 2025 • 0 new comments

• how to solve this problem

  #50721 commented on Mar 27, 2025 • 0 new comments

• [distributed debugger] exception in regular remote worker function leading to access violation when debugger connects

  #51010 commented on Mar 27, 2025 • 0 new comments

• raylet exited immediately because dashboard agent fialed

  #49162 commented on Mar 27, 2025 • 0 new comments

• RLlib: beta1 as a Tensor is not supported for capturable=False and foreach=True

  #51560 commented on Mar 27, 2025 • 0 new comments

• [2/2] [API] Event Observability, add a new event table

  #38708 commented on Mar 26, 2025 • 0 new comments

• [ci][release][core] rewrite RuntimeEnvAgentClient with reusable TCP connection, also test_many_runtime_envs.py with env vars

  #38772 commented on Mar 26, 2025 • 0 new comments

• [TEST] DEBUG

  #38798 commented on Mar 26, 2025 • 0 new comments

• [Data] Add `read_delta` API to read Delta format files

  #38813 commented on Mar 26, 2025 • 0 new comments

• [Core][Label scheduling 8/n]Add length and illegal letters validation to the node labels

  #38824 commented on Mar 26, 2025 • 0 new comments

• Another debug

  #38842 commented on Mar 26, 2025 • 0 new comments

• [Serve] Add more log in the router init step

  #38933 commented on Mar 27, 2025 • 0 new comments

• updates to setup-dev.py to work around the types.py import issues

  #38948 commented on Mar 27, 2025 • 0 new comments

• [WIP][Core] Unflake actor-cancel-test

  #38975 commented on Mar 27, 2025 • 0 new comments

• [WIP] Streaming Generator + actor task lineage reconstruction

  #38982 commented on Mar 27, 2025 • 0 new comments

• [Test] Fix torch dist nccl test

  #38986 commented on Mar 26, 2025 • 0 new comments

• [experiment] rewrite PythonGcsClient with GcsClient

  #39010 commented on Mar 26, 2025 • 0 new comments

• [spark] Improve Ray node memory config calculation logic

  #39149 commented on Mar 27, 2025 • 0 new comments

• [RLlib] Support terminated and truncated in ExternalMultiAgentEnv

  #39175 commented on Mar 27, 2025 • 0 new comments

• [RLlib] DreamerV3: Fix restore from checkpoint functionality

  #39209 commented on Mar 26, 2025 • 0 new comments

• chore: update stale link and comment in tracing_helper.py

  #39239 commented on Mar 27, 2025 • 0 new comments

• Yuming test

  #39242 commented on Mar 26, 2025 • 0 new comments

• [Core] Fix get_next_unordered and get_next

  #39250 commented on Mar 26, 2025 • 0 new comments

• [Core][Observability] Add the scheduling_strategy field to the ActorInfo for the "get actor info" API

  #39256 commented on Mar 26, 2025 • 0 new comments

• Test p

  #39297 commented on Mar 26, 2025 • 0 new comments

• fix typos in router.py

  #39301 commented on Mar 26, 2025 • 0 new comments

• [templates/04_finetuning_llms_with_deepspeed] pin transformers to 4.31.0

  #39372 commented on Mar 26, 2025 • 0 new comments

• [Serve][Debug] websocket test

  #39389 commented on Mar 26, 2025 • 0 new comments

• Update pettingzoo_env.py

  #39431 commented on Mar 26, 2025 • 0 new comments

• [WIP][Core]Add batch remote api for batch submit actor task

  #35597 commented on Mar 26, 2025 • 0 new comments

• [Tune] Add optimizer kwargs for `SkOptSearch`

  #36041 commented on Mar 26, 2025 • 0 new comments

• [autoscaler v2][6/n] introduce instance manager

  #36066 commented on Mar 26, 2025 • 0 new comments

• add setting s3 endpoint-url via env var

  #36114 commented on Mar 27, 2025 • 0 new comments

• Revert "Revert "[Core] Support Arrow zerocopy serialization in object…

  #36153 commented on Mar 27, 2025 • 0 new comments

• [RLlib] Update `check_env` in env.py

  #36463 commented on Mar 27, 2025 • 0 new comments

• [build_base] coroutine cpp

  #36513 commented on Mar 26, 2025 • 0 new comments

• [ci] remove is_automated_build in setup.py

  #36547 commented on Mar 26, 2025 • 0 new comments

• [RLlib] fix PPOConfig warning

  #36595 commented on Mar 27, 2025 • 0 new comments

• [RLlib] fix custom policy examples

  #36600 commented on Mar 27, 2025 • 0 new comments

• [RLlib] Fix A3C use_critic in `rllib_contrib`

  #36613 commented on Mar 27, 2025 • 0 new comments

• WIP python protobuf removal

  #36856 commented on Mar 26, 2025 • 0 new comments

• Add Bazel Steward for dependency management

  #36863 commented on Mar 26, 2025 • 0 new comments

• [Runtime Env] working dir refactor

  #36953 commented on Mar 26, 2025 • 0 new comments

• revert fix: pin libffi=3.3 for base-deps #33294

  #37088 commented on Mar 26, 2025 • 0 new comments

• [core] Remove grpc ClientCallTag

  #37140 commented on Mar 26, 2025 • 0 new comments

• [dag] Show both lib dependency installation instructions on import failure.

  #37236 commented on Mar 27, 2025 • 0 new comments

• obj scale down

  #37687 commented on Mar 26, 2025 • 0 new comments

• [Core] Network benchmark ip

  #37810 commented on Mar 27, 2025 • 0 new comments

• [Data] Add support for ORC format

  #37891 commented on Mar 27, 2025 • 0 new comments

• Enable mixed docker + non-docker clusters

  #37968 commented on Mar 26, 2025 • 0 new comments

• [autoscaler] Use `bash` instead of `/bin/bash`

  #38105 commented on Mar 26, 2025 • 0 new comments

• [tune] Fix error when move file with different disk types

  #38403 commented on Mar 27, 2025 • 0 new comments

• Add Apple silicon GPU(mps) support to ray

  #38464 commented on Mar 26, 2025 • 0 new comments

• fix

  #38623 commented on Mar 26, 2025 • 0 new comments