Skip to content

[CI] Pin mujoco<3.5 to fix CI after mujoco 3.5.0 release#3506

Merged
vmoens merged 1 commit intomainfrom
ci/pin-mujoco-lt-3.5
Feb 13, 2026
Merged

[CI] Pin mujoco<3.5 to fix CI after mujoco 3.5.0 release#3506
vmoens merged 1 commit intomainfrom
ci/pin-mujoco-lt-3.5

Conversation

@vmoens
Copy link
Copy Markdown
Collaborator

@vmoens vmoens commented Feb 13, 2026

Summary

  • MuJoCo 3.5.0 was released on Feb 13, 2026 but dm_control 1.0.36 hasn't been updated to support it yet (latest supports up to MuJoCo 3.4.0).
  • This causes all CI jobs that install mujoco and dm_control to fail.
  • Pin mujoco<3.5 across all CI scripts, workflows, and docs requirements until dm_control releases a compatible version.

Files changed

  • .github/unittest/linux/scripts/run_all.shmujoco>=3.3.7mujoco>=3.3.7,<3.5
  • .github/workflows/benchmarks.ymlmujoco"mujoco<3.5"
  • .github/workflows/benchmarks_pr.ymlmujoco"mujoco<3.5"
  • .github/unittest/linux_libs/scripts_gym/run_all.shmujoco"mujoco<3.5" (2 places)
  • .github/unittest/tutorials/scripts/run_all.sh"mujoco<3.3.6""mujoco<3.5"
  • .github/workflows/test-linux-tutorials.yml"mujoco<3.3.6""mujoco<3.5"
  • docs/requirements.txtmujoco>=3.0.0,<3.3.6mujoco>=3.0.0,<3.5

Test plan

  • CI jobs pass with mujoco pinned to <3.5
  • Remove the pin once dm_control releases a version compatible with mujoco 3.5.0

Made with Cursor

mujoco 3.5.0 was released on Feb 13, 2026 but dm_control 1.0.36 has
not been updated to support it yet (latest supports mujoco 3.4.0).
Pin mujoco<3.5 across all CI scripts, workflows, and docs requirements
until dm_control releases a compatible version.

Co-authored-by: Cursor <cursoragent@cursor.com>
@pytorch-bot
Copy link
Copy Markdown

pytorch-bot bot commented Feb 13, 2026

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3506

Note: Links to docs will display an error until the docs builds have been completed.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 13, 2026
@github-actions github-actions bot added CI Has to do with CI setup (e.g. wheels & builds, tests...) Documentation Improvements or additions to documentation labels Feb 13, 2026
@vmoens vmoens merged commit 05212b8 into main Feb 13, 2026
97 of 117 checks passed
@github-actions
Copy link
Copy Markdown
Contributor

$\color{#D29922}\textsf{\Large&amp;#x26A0;\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 173. Improved: $\large\color{#35bf28}44$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 81.3253μs 79.8738μs 12.5198 KOps/s 12.5527 KOps/s $\color{#d91a1a}-0.26\%$
test_tensor_to_bytestream_speed[torch.save] 0.1387ms 0.1381ms 7.2424 KOps/s 7.3091 KOps/s $\color{#d91a1a}-0.91\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1230s 0.1225s 8.1627 Ops/s 8.9430 Ops/s $\textbf{\color{#d91a1a}-8.73\%}$
test_tensor_to_bytestream_speed[numpy] 2.6123μs 2.5883μs 386.3536 KOps/s 388.1543 KOps/s $\color{#d91a1a}-0.46\%$
test_tensor_to_bytestream_speed[safetensors] 36.7532μs 36.5925μs 27.3280 KOps/s 27.6982 KOps/s $\color{#d91a1a}-1.34\%$
test_simple 0.5458s 0.5448s 1.8354 Ops/s 1.7552 Ops/s $\color{#35bf28}+4.57\%$
test_transformed 1.0929s 1.0919s 0.9158 Ops/s 0.8974 Ops/s $\color{#35bf28}+2.06\%$
test_serial 1.6702s 1.6682s 0.5994 Ops/s 0.5936 Ops/s $\color{#35bf28}+0.98\%$
test_parallel 1.1304s 1.0381s 0.9633 Ops/s 0.9424 Ops/s $\color{#35bf28}+2.22\%$
test_step_mdp_speed[True-True-True-True-True] 0.1982ms 43.2976μs 23.0960 KOps/s 19.9466 KOps/s $\textbf{\color{#35bf28}+15.79\%}$
test_step_mdp_speed[True-True-True-True-False] 57.4610μs 24.9007μs 40.1595 KOps/s 37.8695 KOps/s $\textbf{\color{#35bf28}+6.05\%}$
test_step_mdp_speed[True-True-True-False-True] 60.0010μs 24.7155μs 40.4605 KOps/s 35.3080 KOps/s $\textbf{\color{#35bf28}+14.59\%}$
test_step_mdp_speed[True-True-True-False-False] 40.9510μs 13.8866μs 72.0119 KOps/s 70.5976 KOps/s $\color{#35bf28}+2.00\%$
test_step_mdp_speed[True-True-False-True-True] 87.1110μs 47.3704μs 21.1102 KOps/s 18.9814 KOps/s $\textbf{\color{#35bf28}+11.22\%}$
test_step_mdp_speed[True-True-False-True-False] 57.0210μs 27.7110μs 36.0868 KOps/s 34.7138 KOps/s $\color{#35bf28}+3.96\%$
test_step_mdp_speed[True-True-False-False-True] 61.4210μs 27.9287μs 35.8055 KOps/s 33.8389 KOps/s $\textbf{\color{#35bf28}+5.81\%}$
test_step_mdp_speed[True-True-False-False-False] 45.2210μs 16.5794μs 60.3157 KOps/s 59.1632 KOps/s $\color{#35bf28}+1.95\%$
test_step_mdp_speed[True-False-True-True-True] 87.8720μs 49.8446μs 20.0624 KOps/s 17.9053 KOps/s $\textbf{\color{#35bf28}+12.05\%}$
test_step_mdp_speed[True-False-True-True-False] 63.8710μs 30.4322μs 32.8599 KOps/s 31.2075 KOps/s $\textbf{\color{#35bf28}+5.30\%}$
test_step_mdp_speed[True-False-True-False-True] 54.5310μs 27.4383μs 36.4454 KOps/s 32.1427 KOps/s $\textbf{\color{#35bf28}+13.39\%}$
test_step_mdp_speed[True-False-True-False-False] 40.9310μs 16.6000μs 60.2408 KOps/s 58.8978 KOps/s $\color{#35bf28}+2.28\%$
test_step_mdp_speed[True-False-False-True-True] 85.2110μs 52.8439μs 18.9236 KOps/s 17.2011 KOps/s $\textbf{\color{#35bf28}+10.01\%}$
test_step_mdp_speed[True-False-False-True-False] 69.3720μs 33.6395μs 29.7270 KOps/s 29.2840 KOps/s $\color{#35bf28}+1.51\%$
test_step_mdp_speed[True-False-False-False-True] 62.1110μs 30.2223μs 33.0881 KOps/s 32.4804 KOps/s $\color{#35bf28}+1.87\%$
test_step_mdp_speed[True-False-False-False-False] 61.8510μs 19.2622μs 51.9153 KOps/s 51.2655 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[False-True-True-True-True] 91.9320μs 50.3470μs 19.8621 KOps/s 18.1899 KOps/s $\textbf{\color{#35bf28}+9.19\%}$
test_step_mdp_speed[False-True-True-True-False] 70.4920μs 30.7426μs 32.5281 KOps/s 31.9064 KOps/s $\color{#35bf28}+1.95\%$
test_step_mdp_speed[False-True-True-False-True] 2.3218ms 31.6522μs 31.5933 KOps/s 31.4181 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-True-True-False-False] 55.2110μs 18.5274μs 53.9740 KOps/s 53.0662 KOps/s $\color{#35bf28}+1.71\%$
test_step_mdp_speed[False-True-False-True-True] 0.1091ms 52.8056μs 18.9374 KOps/s 17.2857 KOps/s $\textbf{\color{#35bf28}+9.56\%}$
test_step_mdp_speed[False-True-False-True-False] 76.2420μs 33.3839μs 29.9546 KOps/s 28.9229 KOps/s $\color{#35bf28}+3.57\%$
test_step_mdp_speed[False-True-False-False-True] 74.3320μs 34.1583μs 29.2755 KOps/s 28.7010 KOps/s $\color{#35bf28}+2.00\%$
test_step_mdp_speed[False-True-False-False-False] 55.2710μs 21.0352μs 47.5393 KOps/s 47.1861 KOps/s $\color{#35bf28}+0.75\%$
test_step_mdp_speed[False-False-True-True-True] 96.4720μs 55.2735μs 18.0918 KOps/s 16.9185 KOps/s $\textbf{\color{#35bf28}+6.94\%}$
test_step_mdp_speed[False-False-True-True-False] 64.2320μs 35.8056μs 27.9286 KOps/s 26.7116 KOps/s $\color{#35bf28}+4.56\%$
test_step_mdp_speed[False-False-True-False-True] 67.2820μs 33.7118μs 29.6632 KOps/s 27.6094 KOps/s $\textbf{\color{#35bf28}+7.44\%}$
test_step_mdp_speed[False-False-True-False-False] 53.2510μs 21.1721μs 47.2320 KOps/s 47.0349 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[False-False-False-True-True] 92.1820μs 57.6815μs 17.3366 KOps/s 16.0450 KOps/s $\textbf{\color{#35bf28}+8.05\%}$
test_step_mdp_speed[False-False-False-True-False] 70.7510μs 39.0054μs 25.6375 KOps/s 25.3794 KOps/s $\color{#35bf28}+1.02\%$
test_step_mdp_speed[False-False-False-False-True] 74.5920μs 36.4306μs 27.4495 KOps/s 25.5316 KOps/s $\textbf{\color{#35bf28}+7.51\%}$
test_step_mdp_speed[False-False-False-False-False] 52.0720μs 23.1754μs 43.1493 KOps/s 41.6814 KOps/s $\color{#35bf28}+3.52\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.7278s 0.7255s 1.3784 Ops/s 1.3255 Ops/s $\color{#35bf28}+3.99\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7085s 0.6114s 1.6357 Ops/s 1.6368 Ops/s $\color{#d91a1a}-0.07\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7257s 1.6460s 0.6075 Ops/s 0.6088 Ops/s $\color{#d91a1a}-0.21\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.4998s 1.4216s 0.7034 Ops/s 0.7049 Ops/s $\color{#d91a1a}-0.21\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9752s 1.9016s 0.5259 Ops/s 0.5290 Ops/s $\color{#d91a1a}-0.59\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7507s 1.6743s 0.5973 Ops/s 0.5980 Ops/s $\color{#d91a1a}-0.12\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.6821s 4.6424s 0.2154 Ops/s 0.2146 Ops/s $\color{#35bf28}+0.39\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.5794s 4.4526s 0.2246 Ops/s 0.2256 Ops/s $\color{#d91a1a}-0.46\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9692s 1.8788s 0.5323 Ops/s 0.5230 Ops/s $\color{#35bf28}+1.77\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.6529s 1.5801s 0.6329 Ops/s 0.6283 Ops/s $\color{#35bf28}+0.73\%$
test_values[generalized_advantage_estimate-True-True] 10.5222ms 10.3571ms 96.5521 Ops/s 97.7165 Ops/s $\color{#d91a1a}-1.19\%$
test_values[vec_generalized_advantage_estimate-True-True] 15.0761ms 11.0326ms 90.6405 Ops/s 56.8994 Ops/s $\textbf{\color{#35bf28}+59.30\%}$
test_values[td0_return_estimate-False-False] 0.2225ms 0.1310ms 7.6357 KOps/s 5.2165 KOps/s $\textbf{\color{#35bf28}+46.38\%}$
test_values[td1_return_estimate-False-False] 28.5532ms 27.9053ms 35.8355 Ops/s 36.0040 Ops/s $\color{#d91a1a}-0.47\%$
test_values[vec_td1_return_estimate-False-False] 12.7144ms 11.1287ms 89.8574 Ops/s 55.3344 Ops/s $\textbf{\color{#35bf28}+62.39\%}$
test_values[td_lambda_return_estimate-True-False] 60.1572ms 41.7420ms 23.9567 Ops/s 24.2585 Ops/s $\color{#d91a1a}-1.24\%$
test_values[vec_td_lambda_return_estimate-True-False] 11.6301ms 11.0391ms 90.5871 Ops/s 56.4480 Ops/s $\textbf{\color{#35bf28}+60.48\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.1868ms 9.0861ms 110.0581 Ops/s 110.0238 Ops/s $\color{#35bf28}+0.03\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7427ms 1.4673ms 681.5371 Ops/s 640.1140 Ops/s $\textbf{\color{#35bf28}+6.47\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4688ms 0.4186ms 2.3890 KOps/s 2.3293 KOps/s $\color{#35bf28}+2.56\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.6701ms 29.5801ms 33.8065 Ops/s 28.8240 Ops/s $\textbf{\color{#35bf28}+17.29\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.7980ms 1.7017ms 587.6439 Ops/s 587.1371 Ops/s $\color{#35bf28}+0.09\%$
test_dqn_speed[False-None] 1.5067ms 1.4005ms 714.0475 Ops/s 698.2103 Ops/s $\color{#35bf28}+2.27\%$
test_dqn_speed[False-backward] 1.9742ms 1.9222ms 520.2324 Ops/s 504.0655 Ops/s $\color{#35bf28}+3.21\%$
test_dqn_speed[True-None] 0.8190ms 0.5754ms 1.7380 KOps/s 1.7507 KOps/s $\color{#d91a1a}-0.73\%$
test_dqn_speed[True-backward] 1.1321ms 1.0115ms 988.6305 Ops/s 823.7654 Ops/s $\textbf{\color{#35bf28}+20.01\%}$
test_dqn_speed[reduce-overhead-None] 0.6639ms 0.5330ms 1.8761 KOps/s 1.7057 KOps/s $\textbf{\color{#35bf28}+9.99\%}$
test_ddpg_speed[False-None] 3.2471ms 2.8600ms 349.6477 Ops/s 346.1813 Ops/s $\color{#35bf28}+1.00\%$
test_ddpg_speed[False-backward] 4.1792ms 4.0987ms 243.9772 Ops/s 240.3445 Ops/s $\color{#35bf28}+1.51\%$
test_ddpg_speed[True-None] 1.6290ms 1.3903ms 719.2864 Ops/s 686.0172 Ops/s $\color{#35bf28}+4.85\%$
test_ddpg_speed[True-backward] 2.4242ms 2.3839ms 419.4881 Ops/s 406.7106 Ops/s $\color{#35bf28}+3.14\%$
test_ddpg_speed[reduce-overhead-None] 1.5870ms 1.3962ms 716.2210 Ops/s 702.1291 Ops/s $\color{#35bf28}+2.01\%$
test_sac_speed[False-None] 8.6304ms 7.9908ms 125.1432 Ops/s 122.9127 Ops/s $\color{#35bf28}+1.81\%$
test_sac_speed[False-backward] 11.8615ms 11.3012ms 88.4866 Ops/s 87.1210 Ops/s $\color{#35bf28}+1.57\%$
test_sac_speed[True-None] 2.2873ms 2.1597ms 463.0234 Ops/s 457.5896 Ops/s $\color{#35bf28}+1.19\%$
test_sac_speed[True-backward] 4.1274ms 4.0411ms 247.4562 Ops/s 245.2613 Ops/s $\color{#35bf28}+0.89\%$
test_sac_speed[reduce-overhead-None] 2.3164ms 2.1409ms 467.0915 Ops/s 460.3190 Ops/s $\color{#35bf28}+1.47\%$
test_redq_speed[False-None] 11.1098ms 10.3780ms 96.3577 Ops/s 94.6381 Ops/s $\color{#35bf28}+1.82\%$
test_redq_speed[False-backward] 18.7348ms 17.6908ms 56.5265 Ops/s 56.5377 Ops/s $\color{#d91a1a}-0.02\%$
test_redq_speed[True-None] 4.6422ms 4.4232ms 226.0830 Ops/s 228.6001 Ops/s $\color{#d91a1a}-1.10\%$
test_redq_speed[True-backward] 10.1726ms 9.8719ms 101.2979 Ops/s 104.2746 Ops/s $\color{#d91a1a}-2.85\%$
test_redq_speed[reduce-overhead-None] 4.8566ms 4.4682ms 223.8058 Ops/s 223.5050 Ops/s $\color{#35bf28}+0.13\%$
test_redq_deprec_speed[False-None] 11.6210ms 11.1347ms 89.8096 Ops/s 88.0601 Ops/s $\color{#35bf28}+1.99\%$
test_redq_deprec_speed[False-backward] 16.3715ms 15.9209ms 62.8105 Ops/s 62.4032 Ops/s $\color{#35bf28}+0.65\%$
test_redq_deprec_speed[True-None] 3.9241ms 3.6773ms 271.9356 Ops/s 274.7301 Ops/s $\color{#d91a1a}-1.02\%$
test_redq_deprec_speed[True-backward] 7.9035ms 7.6607ms 130.5359 Ops/s 121.2775 Ops/s $\textbf{\color{#35bf28}+7.63\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8597ms 3.5927ms 278.3457 Ops/s 267.8257 Ops/s $\color{#35bf28}+3.93\%$
test_td3_speed[False-None] 8.1314ms 7.9950ms 125.0786 Ops/s 122.2227 Ops/s $\color{#35bf28}+2.34\%$
test_td3_speed[False-backward] 11.3260ms 10.8886ms 91.8392 Ops/s 89.4725 Ops/s $\color{#35bf28}+2.65\%$
test_td3_speed[True-None] 1.8390ms 1.8210ms 549.1477 Ops/s 534.9727 Ops/s $\color{#35bf28}+2.65\%$
test_td3_speed[True-backward] 3.7974ms 3.6191ms 276.3121 Ops/s 240.0857 Ops/s $\textbf{\color{#35bf28}+15.09\%}$
test_td3_speed[reduce-overhead-None] 1.8202ms 1.7880ms 559.2798 Ops/s 540.7884 Ops/s $\color{#35bf28}+3.42\%$
test_cql_speed[False-None] 28.5927ms 26.0732ms 38.3536 Ops/s 37.3754 Ops/s $\color{#35bf28}+2.62\%$
test_cql_speed[False-backward] 36.0530ms 35.1811ms 28.4243 Ops/s 27.0560 Ops/s $\textbf{\color{#35bf28}+5.06\%}$
test_cql_speed[True-None] 12.7616ms 12.4072ms 80.5982 Ops/s 80.4878 Ops/s $\color{#35bf28}+0.14\%$
test_cql_speed[True-backward] 18.8845ms 18.5902ms 53.7919 Ops/s 55.3899 Ops/s $\color{#d91a1a}-2.88\%$
test_cql_speed[reduce-overhead-None] 13.0699ms 12.4983ms 80.0111 Ops/s 77.4208 Ops/s $\color{#35bf28}+3.35\%$
test_a2c_speed[False-None] 5.6759ms 5.4272ms 184.2578 Ops/s 166.1811 Ops/s $\textbf{\color{#35bf28}+10.88\%}$
test_a2c_speed[False-backward] 12.0693ms 11.8631ms 84.2953 Ops/s 79.8507 Ops/s $\textbf{\color{#35bf28}+5.57\%}$
test_a2c_speed[True-None] 3.8440ms 3.7308ms 268.0366 Ops/s 259.5936 Ops/s $\color{#35bf28}+3.25\%$
test_a2c_speed[True-backward] 8.7172ms 8.5347ms 117.1686 Ops/s 112.7062 Ops/s $\color{#35bf28}+3.96\%$
test_a2c_speed[reduce-overhead-None] 4.1332ms 3.7339ms 267.8146 Ops/s 266.6351 Ops/s $\color{#35bf28}+0.44\%$
test_ppo_speed[False-None] 6.1116ms 5.9535ms 167.9698 Ops/s 151.4594 Ops/s $\textbf{\color{#35bf28}+10.90\%}$
test_ppo_speed[False-backward] 13.0248ms 12.6692ms 78.9318 Ops/s 76.4285 Ops/s $\color{#35bf28}+3.28\%$
test_ppo_speed[True-None] 3.8068ms 3.6553ms 273.5764 Ops/s 270.2580 Ops/s $\color{#35bf28}+1.23\%$
test_ppo_speed[True-backward] 8.6599ms 8.4459ms 118.4008 Ops/s 117.3240 Ops/s $\color{#35bf28}+0.92\%$
test_ppo_speed[reduce-overhead-None] 4.3330ms 3.6393ms 274.7799 Ops/s 272.8229 Ops/s $\color{#35bf28}+0.72\%$
test_reinforce_speed[False-None] 4.8867ms 4.5916ms 217.7898 Ops/s 192.8114 Ops/s $\textbf{\color{#35bf28}+12.95\%}$
test_reinforce_speed[False-backward] 7.6956ms 7.4452ms 134.3154 Ops/s 124.4057 Ops/s $\textbf{\color{#35bf28}+7.97\%}$
test_reinforce_speed[True-None] 3.1978ms 2.9185ms 342.6475 Ops/s 337.6308 Ops/s $\color{#35bf28}+1.49\%$
test_reinforce_speed[True-backward] 8.0560ms 7.7237ms 129.4715 Ops/s 125.5995 Ops/s $\color{#35bf28}+3.08\%$
test_reinforce_speed[reduce-overhead-None] 3.0659ms 2.8757ms 347.7442 Ops/s 345.4238 Ops/s $\color{#35bf28}+0.67\%$
test_iql_speed[False-None] 26.3561ms 20.3898ms 49.0442 Ops/s 46.3928 Ops/s $\textbf{\color{#35bf28}+5.71\%}$
test_iql_speed[False-backward] 35.2902ms 30.7503ms 32.5200 Ops/s 31.3458 Ops/s $\color{#35bf28}+3.75\%$
test_iql_speed[True-None] 8.8289ms 8.5053ms 117.5734 Ops/s 113.0150 Ops/s $\color{#35bf28}+4.03\%$
test_iql_speed[True-backward] 16.9624ms 16.6613ms 60.0192 Ops/s 59.3016 Ops/s $\color{#35bf28}+1.21\%$
test_iql_speed[reduce-overhead-None] 8.8534ms 8.5870ms 116.4551 Ops/s 115.1880 Ops/s $\color{#35bf28}+1.10\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1800ms 6.0260ms 165.9484 Ops/s 163.8754 Ops/s $\color{#35bf28}+1.27\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0820ms 0.3281ms 3.0480 KOps/s 2.6641 KOps/s $\textbf{\color{#35bf28}+14.41\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5074ms 0.2659ms 3.7613 KOps/s 2.7923 KOps/s $\textbf{\color{#35bf28}+34.70\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1335ms 5.8298ms 171.5317 Ops/s 172.9495 Ops/s $\color{#d91a1a}-0.82\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8673ms 0.3051ms 3.2778 KOps/s 2.7796 KOps/s $\textbf{\color{#35bf28}+17.92\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4809ms 0.2588ms 3.8634 KOps/s 2.8734 KOps/s $\textbf{\color{#35bf28}+34.45\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6518ms 1.2819ms 780.0643 Ops/s 712.1755 Ops/s $\textbf{\color{#35bf28}+9.53\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6792ms 1.2114ms 825.5055 Ops/s 750.2126 Ops/s $\textbf{\color{#35bf28}+10.04\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.9374ms 6.1036ms 163.8385 Ops/s 167.9241 Ops/s $\color{#d91a1a}-2.43\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9187ms 0.4322ms 2.3135 KOps/s 2.0581 KOps/s $\textbf{\color{#35bf28}+12.41\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6594ms 0.4126ms 2.4238 KOps/s 2.2694 KOps/s $\textbf{\color{#35bf28}+6.80\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0660ms 5.9016ms 169.4447 Ops/s 169.6263 Ops/s $\color{#d91a1a}-0.11\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.7362ms 0.3427ms 2.9182 KOps/s 2.9221 KOps/s $\color{#d91a1a}-0.14\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5654ms 0.3380ms 2.9584 KOps/s 2.8485 KOps/s $\color{#35bf28}+3.86\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0743ms 5.8312ms 171.4905 Ops/s 171.9436 Ops/s $\color{#d91a1a}-0.26\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9727ms 0.4033ms 2.4794 KOps/s 2.7561 KOps/s $\textbf{\color{#d91a1a}-10.04\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5617ms 0.3435ms 2.9110 KOps/s 3.1407 KOps/s $\textbf{\color{#d91a1a}-7.31\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.8084ms 6.0377ms 165.6255 Ops/s 169.1394 Ops/s $\color{#d91a1a}-2.08\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8497ms 0.4699ms 2.1280 KOps/s 1.9036 KOps/s $\textbf{\color{#35bf28}+11.79\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7437ms 0.4570ms 2.1880 KOps/s 2.1812 KOps/s $\color{#35bf28}+0.31\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.6548s 18.0210ms 55.4909 Ops/s 197.9904 Ops/s $\textbf{\color{#d91a1a}-71.97\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 5.0389ms 1.8344ms 545.1286 Ops/s 68.7492 Ops/s $\textbf{\color{#35bf28}+692.92\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.4595ms 1.1095ms 901.2720 Ops/s 1.0412 KOps/s $\textbf{\color{#d91a1a}-13.44\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 9.2199ms 5.0717ms 197.1735 Ops/s 194.9708 Ops/s $\color{#35bf28}+1.13\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.9404ms 1.7573ms 569.0491 Ops/s 510.6130 Ops/s $\textbf{\color{#35bf28}+11.44\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 12.2728ms 1.3174ms 759.0619 Ops/s 1.1320 KOps/s $\textbf{\color{#d91a1a}-32.95\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5190s 15.4928ms 64.5461 Ops/s 187.2194 Ops/s $\textbf{\color{#d91a1a}-65.52\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.0057ms 1.8980ms 526.8782 Ops/s 514.6659 Ops/s $\color{#35bf28}+2.37\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.8837ms 1.0570ms 946.0514 Ops/s 774.5855 Ops/s $\textbf{\color{#35bf28}+22.14\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.8212ms 36.1521ms 27.6609 Ops/s 27.3394 Ops/s $\color{#35bf28}+1.18\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.4685ms 18.4950ms 54.0686 Ops/s 54.0904 Ops/s $\color{#d91a1a}-0.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 40.4590ms 37.3355ms 26.7842 Ops/s 26.9658 Ops/s $\color{#d91a1a}-0.67\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.6417ms 18.8068ms 53.1723 Ops/s 53.8932 Ops/s $\color{#d91a1a}-1.34\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 42.3848ms 39.1432ms 25.5472 Ops/s 25.7151 Ops/s $\color{#d91a1a}-0.65\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.9714ms 20.5300ms 48.7093 Ops/s 50.1011 Ops/s $\color{#d91a1a}-2.78\%$
test_storage_write_lazystack[50-img_shape0-small] 0.8548ms 0.2263ms 4.4189 KOps/s 4.4603 KOps/s $\color{#d91a1a}-0.93\%$
test_storage_write_lazystack[100-img_shape1-atari] 1.7370ms 1.4062ms 711.1496 Ops/s 711.6665 Ops/s $\color{#d91a1a}-0.07\%$
test_storage_write_lazystack[100-img_shape2-large_img] 2.4848ms 2.3259ms 429.9444 Ops/s 436.9538 Ops/s $\color{#d91a1a}-1.60\%$
test_storage_write_lazystack[200-img_shape3-large_batch] 3.0781ms 2.8571ms 350.0043 Ops/s 344.0025 Ops/s $\color{#35bf28}+1.74\%$
test_storage_write_contiguous[50-img_shape0-small] 0.2193ms 0.1382ms 7.2356 KOps/s 7.4827 KOps/s $\color{#d91a1a}-3.30\%$
test_storage_write_contiguous[100-img_shape1-atari] 0.3463ms 0.1895ms 5.2783 KOps/s 5.2110 KOps/s $\color{#35bf28}+1.29\%$
test_storage_write_contiguous[100-img_shape2-large_img] 1.8219ms 1.7413ms 574.2946 Ops/s 570.0189 Ops/s $\color{#35bf28}+0.75\%$
test_storage_write_contiguous[200-img_shape3-large_batch] 1.4379ms 1.2927ms 773.5890 Ops/s 771.3798 Ops/s $\color{#35bf28}+0.29\%$
test_collector_stack_then_write[50-img_shape0-small] 1.6533ms 1.1283ms 886.3156 Ops/s 892.5676 Ops/s $\color{#d91a1a}-0.70\%$
test_collector_stack_then_write[100-img_shape1-atari] 3.6676ms 3.5971ms 277.9980 Ops/s 280.4295 Ops/s $\color{#d91a1a}-0.87\%$
test_collector_stack_then_write[100-img_shape2-large_img] 5.8207ms 5.6432ms 177.2045 Ops/s 174.0046 Ops/s $\color{#35bf28}+1.84\%$
test_collector_stack_then_write[200-img_shape3-large_batch] 7.3628ms 7.0279ms 142.2892 Ops/s 137.0890 Ops/s $\color{#35bf28}+3.79\%$
test_collector_lazystack_then_write[50-img_shape0-small] 0.4287ms 0.2722ms 3.6738 KOps/s 3.5031 KOps/s $\color{#35bf28}+4.87\%$
test_collector_lazystack_then_write[100-img_shape1-atari] 1.6904ms 1.4910ms 670.6960 Ops/s 655.6719 Ops/s $\color{#35bf28}+2.29\%$
test_collector_lazystack_then_write[100-img_shape2-large_img] 2.9415ms 2.4324ms 411.1133 Ops/s 411.2626 Ops/s $\color{#d91a1a}-0.04\%$
test_collector_lazystack_then_write[200-img_shape3-large_batch] 3.3879ms 3.0843ms 324.2216 Ops/s 319.4990 Ops/s $\color{#35bf28}+1.48\%$
test_collector_without_rb[100-img_shape0-atari] 33.2377ms 32.8362ms 30.4542 Ops/s 30.0790 Ops/s $\color{#35bf28}+1.25\%$
test_collector_without_rb[200-img_shape1-large_batch] 65.3271ms 65.1868ms 15.3405 Ops/s 15.2397 Ops/s $\color{#35bf28}+0.66\%$
test_collector_with_rb[100-img_shape0-atari] 38.0464ms 37.2897ms 26.8170 Ops/s 26.3630 Ops/s $\color{#35bf28}+1.72\%$
test_collector_with_rb[200-img_shape1-large_batch] 73.7869ms 72.9763ms 13.7031 Ops/s 13.3733 Ops/s $\color{#35bf28}+2.47\%$

@github-actions
Copy link
Copy Markdown
Contributor

$\color{#D29922}\textsf{\Large&amp;#x26A0;\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 172. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 79.8709μs 79.0096μs 12.6567 KOps/s 12.7030 KOps/s $\color{#d91a1a}-0.36\%$
test_tensor_to_bytestream_speed[torch.save] 0.1366ms 0.1361ms 7.3472 KOps/s 7.3516 KOps/s $\color{#d91a1a}-0.06\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1106s 0.1096s 9.1247 Ops/s 9.4465 Ops/s $\color{#d91a1a}-3.41\%$
test_tensor_to_bytestream_speed[numpy] 2.5057μs 2.4999μs 400.0210 KOps/s 412.2077 KOps/s $\color{#d91a1a}-2.96\%$
test_tensor_to_bytestream_speed[safetensors] 39.6431μs 38.0246μs 26.2988 KOps/s 25.8241 KOps/s $\color{#35bf28}+1.84\%$
test_simple 0.7990s 0.7802s 1.2817 Ops/s 1.2475 Ops/s $\color{#35bf28}+2.74\%$
test_transformed 1.4785s 1.3831s 0.7230 Ops/s 0.7195 Ops/s $\color{#35bf28}+0.49\%$
test_serial 2.3665s 2.2785s 0.4389 Ops/s 0.4413 Ops/s $\color{#d91a1a}-0.55\%$
test_parallel 1.8944s 1.8041s 0.5543 Ops/s 0.5613 Ops/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[True-True-True-True-True] 90.1620μs 43.2687μs 23.1114 KOps/s 22.7886 KOps/s $\color{#35bf28}+1.42\%$
test_step_mdp_speed[True-True-True-True-False] 54.6210μs 24.4275μs 40.9375 KOps/s 40.8637 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[True-True-True-False-True] 0.1887ms 23.4927μs 42.5665 KOps/s 40.6919 KOps/s $\color{#35bf28}+4.61\%$
test_step_mdp_speed[True-True-True-False-False] 37.4300μs 13.3827μs 74.7235 KOps/s 73.5068 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[True-True-False-True-True] 73.9110μs 46.5399μs 21.4869 KOps/s 21.7834 KOps/s $\color{#d91a1a}-1.36\%$
test_step_mdp_speed[True-True-False-True-False] 60.3310μs 26.7991μs 37.3147 KOps/s 36.3329 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[True-True-False-False-True] 57.7610μs 27.3003μs 36.6296 KOps/s 36.8684 KOps/s $\color{#d91a1a}-0.65\%$
test_step_mdp_speed[True-True-False-False-False] 47.4900μs 16.3393μs 61.2023 KOps/s 60.3354 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[True-False-True-True-True] 87.4110μs 49.6237μs 20.1517 KOps/s 20.0409 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[True-False-True-True-False] 57.2710μs 29.5704μs 33.8175 KOps/s 32.6589 KOps/s $\color{#35bf28}+3.55\%$
test_step_mdp_speed[True-False-True-False-True] 54.8110μs 26.9361μs 37.1250 KOps/s 36.7922 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[True-False-True-False-False] 51.7610μs 16.1689μs 61.8472 KOps/s 60.5783 KOps/s $\color{#35bf28}+2.09\%$
test_step_mdp_speed[True-False-False-True-True] 89.0910μs 51.6427μs 19.3638 KOps/s 19.2735 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[True-False-False-True-False] 67.5110μs 31.9850μs 31.2646 KOps/s 29.5537 KOps/s $\textbf{\color{#35bf28}+5.79\%}$
test_step_mdp_speed[True-False-False-False-True] 60.3710μs 29.3051μs 34.1238 KOps/s 34.2001 KOps/s $\color{#d91a1a}-0.22\%$
test_step_mdp_speed[True-False-False-False-False] 67.4410μs 18.7300μs 53.3904 KOps/s 52.6228 KOps/s $\color{#35bf28}+1.46\%$
test_step_mdp_speed[False-True-True-True-True] 95.5610μs 49.2825μs 20.2912 KOps/s 20.2899 KOps/s $+0.01\%$
test_step_mdp_speed[False-True-True-True-False] 59.6210μs 30.1882μs 33.1255 KOps/s 32.7574 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[False-True-True-False-True] 2.4050ms 31.6426μs 31.6030 KOps/s 32.4762 KOps/s $\color{#d91a1a}-2.69\%$
test_step_mdp_speed[False-True-True-False-False] 57.6410μs 17.9013μs 55.8617 KOps/s 55.0163 KOps/s $\color{#35bf28}+1.54\%$
test_step_mdp_speed[False-True-False-True-True] 94.6210μs 52.0840μs 19.1998 KOps/s 19.2123 KOps/s $\color{#d91a1a}-0.07\%$
test_step_mdp_speed[False-True-False-True-False] 60.5110μs 32.1494μs 31.1048 KOps/s 30.6814 KOps/s $\color{#35bf28}+1.38\%$
test_step_mdp_speed[False-True-False-False-True] 67.1110μs 33.0190μs 30.2856 KOps/s 30.4087 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[False-True-False-False-False] 47.6800μs 20.0378μs 49.9056 KOps/s 47.9427 KOps/s $\color{#35bf28}+4.09\%$
test_step_mdp_speed[False-False-True-True-True] 87.8720μs 54.0048μs 18.5169 KOps/s 18.7755 KOps/s $\color{#d91a1a}-1.38\%$
test_step_mdp_speed[False-False-True-True-False] 81.0810μs 34.8617μs 28.6847 KOps/s 28.3681 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[False-False-True-False-True] 63.9610μs 32.7254μs 30.5573 KOps/s 30.8247 KOps/s $\color{#d91a1a}-0.87\%$
test_step_mdp_speed[False-False-True-False-False] 52.3510μs 19.8830μs 50.2943 KOps/s 48.5663 KOps/s $\color{#35bf28}+3.56\%$
test_step_mdp_speed[False-False-False-True-True] 93.5720μs 55.0465μs 18.1665 KOps/s 17.8597 KOps/s $\color{#35bf28}+1.72\%$
test_step_mdp_speed[False-False-False-True-False] 71.8310μs 37.8534μs 26.4177 KOps/s 26.5679 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[False-False-False-False-True] 69.1510μs 35.4809μs 28.1842 KOps/s 28.1503 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[False-False-False-False-False] 59.7510μs 22.7531μs 43.9500 KOps/s 43.6299 KOps/s $\color{#35bf28}+0.73\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8299s 0.7300s 1.3698 Ops/s 1.3596 Ops/s $\color{#35bf28}+0.75\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.6911s 0.5942s 1.6830 Ops/s 1.6522 Ops/s $\color{#35bf28}+1.86\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.6798s 1.5977s 0.6259 Ops/s 0.6183 Ops/s $\color{#35bf28}+1.22\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.4588s 1.3767s 0.7264 Ops/s 0.7139 Ops/s $\color{#35bf28}+1.74\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9153s 1.8386s 0.5439 Ops/s 0.5382 Ops/s $\color{#35bf28}+1.05\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.6991s 1.6191s 0.6176 Ops/s 0.6106 Ops/s $\color{#35bf28}+1.15\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.5964s 4.5054s 0.2220 Ops/s 0.2191 Ops/s $\color{#35bf28}+1.31\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4891s 4.4207s 0.2262 Ops/s 0.2297 Ops/s $\color{#d91a1a}-1.51\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9553s 1.8341s 0.5452 Ops/s 0.5338 Ops/s $\color{#35bf28}+2.15\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7228s 1.6168s 0.6185 Ops/s 0.6471 Ops/s $\color{#d91a1a}-4.42\%$
test_values[generalized_advantage_estimate-True-True] 19.7466ms 19.2818ms 51.8624 Ops/s 50.7579 Ops/s $\color{#35bf28}+2.18\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1293s 3.4992ms 285.7823 Ops/s 283.9709 Ops/s $\color{#35bf28}+0.64\%$
test_values[td0_return_estimate-False-False] 0.1055ms 81.5834μs 12.2574 KOps/s 12.1186 KOps/s $\color{#35bf28}+1.14\%$
test_values[td1_return_estimate-False-False] 46.3864ms 45.9084ms 21.7825 Ops/s 21.3310 Ops/s $\color{#35bf28}+2.12\%$
test_values[vec_td1_return_estimate-False-False] 1.3599ms 1.0702ms 934.4449 Ops/s 924.5915 Ops/s $\color{#35bf28}+1.07\%$
test_values[td_lambda_return_estimate-True-False] 76.1028ms 75.4823ms 13.2481 Ops/s 12.9551 Ops/s $\color{#35bf28}+2.26\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2729ms 1.0673ms 936.9815 Ops/s 927.3705 Ops/s $\color{#35bf28}+1.04\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 19.6101ms 19.3241ms 51.7489 Ops/s 48.1684 Ops/s $\textbf{\color{#35bf28}+7.43\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0322ms 0.7378ms 1.3555 KOps/s 1.3515 KOps/s $\color{#35bf28}+0.29\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8050ms 0.6615ms 1.5117 KOps/s 1.4745 KOps/s $\color{#35bf28}+2.52\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5239ms 1.4732ms 678.7935 Ops/s 677.8120 Ops/s $\color{#35bf28}+0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7229ms 0.6789ms 1.4730 KOps/s 1.4273 KOps/s $\color{#35bf28}+3.20\%$
test_dqn_speed[False-None] 1.6068ms 1.4921ms 670.2173 Ops/s 665.9324 Ops/s $\color{#35bf28}+0.64\%$
test_dqn_speed[False-backward] 2.1935ms 2.1258ms 470.4126 Ops/s 467.4964 Ops/s $\color{#35bf28}+0.62\%$
test_dqn_speed[True-None] 0.6841ms 0.5657ms 1.7678 KOps/s 1.7574 KOps/s $\color{#35bf28}+0.59\%$
test_dqn_speed[True-backward] 1.1344ms 1.0969ms 911.6735 Ops/s 835.7211 Ops/s $\textbf{\color{#35bf28}+9.09\%}$
test_dqn_speed[reduce-overhead-None] 1.0161ms 0.5805ms 1.7227 KOps/s 1.6458 KOps/s $\color{#35bf28}+4.68\%$
test_ddpg_speed[False-None] 3.2723ms 2.8285ms 353.5487 Ops/s 355.5290 Ops/s $\color{#d91a1a}-0.56\%$
test_ddpg_speed[False-backward] 4.4977ms 4.0657ms 245.9615 Ops/s 238.7579 Ops/s $\color{#35bf28}+3.02\%$
test_ddpg_speed[True-None] 1.7827ms 1.2988ms 769.9333 Ops/s 765.2132 Ops/s $\color{#35bf28}+0.62\%$
test_ddpg_speed[True-backward] 2.4273ms 2.3412ms 427.1367 Ops/s 397.8778 Ops/s $\textbf{\color{#35bf28}+7.35\%}$
test_ddpg_speed[reduce-overhead-None] 1.8494ms 1.3573ms 736.7512 Ops/s 751.1453 Ops/s $\color{#d91a1a}-1.92\%$
test_sac_speed[False-None] 8.7229ms 8.2134ms 121.7521 Ops/s 123.3460 Ops/s $\color{#d91a1a}-1.29\%$
test_sac_speed[False-backward] 11.4816ms 11.0142ms 90.7917 Ops/s 88.0639 Ops/s $\color{#35bf28}+3.10\%$
test_sac_speed[True-None] 2.3008ms 1.8689ms 535.0782 Ops/s 553.9172 Ops/s $\color{#d91a1a}-3.40\%$
test_sac_speed[True-backward] 3.8935ms 3.4319ms 291.3863 Ops/s 277.6980 Ops/s $\color{#35bf28}+4.93\%$
test_sac_speed[reduce-overhead-None] 19.2660ms 10.9558ms 91.2755 Ops/s 83.1543 Ops/s $\textbf{\color{#35bf28}+9.77\%}$
test_redq_deprec_speed[False-None] 9.9795ms 9.1390ms 109.4213 Ops/s 109.5473 Ops/s $\color{#d91a1a}-0.11\%$
test_redq_deprec_speed[False-backward] 12.9065ms 12.1696ms 82.1718 Ops/s 81.8161 Ops/s $\color{#35bf28}+0.43\%$
test_redq_deprec_speed[True-None] 2.6962ms 2.5195ms 396.9066 Ops/s 397.7030 Ops/s $\color{#d91a1a}-0.20\%$
test_redq_deprec_speed[True-backward] 4.6215ms 4.0721ms 245.5718 Ops/s 242.5649 Ops/s $\color{#35bf28}+1.24\%$
test_redq_deprec_speed[reduce-overhead-None] 16.2564ms 9.8558ms 101.4626 Ops/s 102.2630 Ops/s $\color{#d91a1a}-0.78\%$
test_td3_speed[False-None] 8.1941ms 8.0622ms 124.0359 Ops/s 124.0140 Ops/s $\color{#35bf28}+0.02\%$
test_td3_speed[False-backward] 10.8798ms 10.4272ms 95.9029 Ops/s 94.4398 Ops/s $\color{#35bf28}+1.55\%$
test_td3_speed[True-None] 1.7465ms 1.6826ms 594.3257 Ops/s 618.6584 Ops/s $\color{#d91a1a}-3.93\%$
test_td3_speed[True-backward] 3.1361ms 3.0526ms 327.5931 Ops/s 308.9491 Ops/s $\textbf{\color{#35bf28}+6.03\%}$
test_td3_speed[reduce-overhead-None] 86.9268ms 24.2939ms 41.1626 Ops/s 42.0069 Ops/s $\color{#d91a1a}-2.01\%$
test_cql_speed[False-None] 17.5759ms 16.9398ms 59.0327 Ops/s 58.9501 Ops/s $\color{#35bf28}+0.14\%$
test_cql_speed[False-backward] 23.3871ms 22.0723ms 45.3058 Ops/s 44.5612 Ops/s $\color{#35bf28}+1.67\%$
test_cql_speed[True-None] 3.9782ms 3.2365ms 308.9803 Ops/s 308.1636 Ops/s $\color{#35bf28}+0.27\%$
test_cql_speed[True-backward] 5.8288ms 5.3452ms 187.0834 Ops/s 180.8131 Ops/s $\color{#35bf28}+3.47\%$
test_cql_speed[reduce-overhead-None] 19.0483ms 11.8811ms 84.1672 Ops/s 84.6645 Ops/s $\color{#d91a1a}-0.59\%$
test_a2c_speed[False-None] 3.9487ms 3.1804ms 314.4238 Ops/s 314.7070 Ops/s $\color{#d91a1a}-0.09\%$
test_a2c_speed[False-backward] 6.5051ms 6.0044ms 166.5434 Ops/s 156.1762 Ops/s $\textbf{\color{#35bf28}+6.64\%}$
test_a2c_speed[True-None] 1.8216ms 1.3195ms 757.8609 Ops/s 764.9660 Ops/s $\color{#d91a1a}-0.93\%$
test_a2c_speed[True-backward] 3.0272ms 2.9060ms 344.1150 Ops/s 322.7794 Ops/s $\textbf{\color{#35bf28}+6.61\%}$
test_a2c_speed[reduce-overhead-None] 1.3681ms 0.9558ms 1.0463 KOps/s 1.0200 KOps/s $\color{#35bf28}+2.58\%$
test_ppo_speed[False-None] 4.1934ms 3.7808ms 264.4952 Ops/s 263.7901 Ops/s $\color{#35bf28}+0.27\%$
test_ppo_speed[False-backward] 7.4988ms 6.9011ms 144.9038 Ops/s 141.3061 Ops/s $\color{#35bf28}+2.55\%$
test_ppo_speed[True-None] 1.5145ms 1.4071ms 710.6595 Ops/s 711.1110 Ops/s $\color{#d91a1a}-0.06\%$
test_ppo_speed[True-backward] 3.2830ms 3.1619ms 316.2689 Ops/s 307.8368 Ops/s $\color{#35bf28}+2.74\%$
test_ppo_speed[reduce-overhead-None] 1.2236ms 1.0261ms 974.5263 Ops/s 930.7899 Ops/s $\color{#35bf28}+4.70\%$
test_reinforce_speed[False-None] 2.3716ms 2.2525ms 443.9518 Ops/s 441.9049 Ops/s $\color{#35bf28}+0.46\%$
test_reinforce_speed[False-backward] 3.7475ms 3.3814ms 295.7361 Ops/s 305.2190 Ops/s $\color{#d91a1a}-3.11\%$
test_reinforce_speed[True-None] 1.4118ms 1.2820ms 780.0246 Ops/s 792.6238 Ops/s $\color{#d91a1a}-1.59\%$
test_reinforce_speed[True-backward] 3.0775ms 2.8891ms 346.1257 Ops/s 328.5022 Ops/s $\textbf{\color{#35bf28}+5.36\%}$
test_reinforce_speed[reduce-overhead-None] 0.4427s 10.3745ms 96.3899 Ops/s 105.9132 Ops/s $\textbf{\color{#d91a1a}-8.99\%}$
test_iql_speed[False-None] 10.1460ms 9.3803ms 106.6060 Ops/s 107.9441 Ops/s $\color{#d91a1a}-1.24\%$
test_iql_speed[False-backward] 13.6240ms 12.9774ms 77.0569 Ops/s 75.8521 Ops/s $\color{#35bf28}+1.59\%$
test_iql_speed[True-None] 2.4993ms 2.1955ms 455.4790 Ops/s 461.9774 Ops/s $\color{#d91a1a}-1.41\%$
test_iql_speed[True-backward] 4.8968ms 4.6671ms 214.2662 Ops/s 204.8453 Ops/s $\color{#35bf28}+4.60\%$
test_iql_speed[reduce-overhead-None] 18.1206ms 10.6403ms 93.9824 Ops/s 94.8300 Ops/s $\color{#d91a1a}-0.89\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2840ms 5.7657ms 173.4388 Ops/s 175.5299 Ops/s $\color{#d91a1a}-1.19\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2360ms 0.2772ms 3.6070 KOps/s 2.9044 KOps/s $\textbf{\color{#35bf28}+24.19\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4997ms 0.2584ms 3.8695 KOps/s 2.8642 KOps/s $\textbf{\color{#35bf28}+35.10\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7854ms 5.5284ms 180.8848 Ops/s 179.6986 Ops/s $\color{#35bf28}+0.66\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9782ms 0.3320ms 3.0117 KOps/s 3.6554 KOps/s $\textbf{\color{#d91a1a}-17.61\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5473ms 0.3211ms 3.1138 KOps/s 3.8844 KOps/s $\textbf{\color{#d91a1a}-19.84\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5092ms 1.2964ms 771.3401 Ops/s 823.3654 Ops/s $\textbf{\color{#d91a1a}-6.32\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4933ms 1.2204ms 819.4336 Ops/s 879.4692 Ops/s $\textbf{\color{#d91a1a}-6.83\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8487ms 5.6669ms 176.4636 Ops/s 172.1159 Ops/s $\color{#35bf28}+2.53\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8034ms 0.4842ms 2.0651 KOps/s 1.9897 KOps/s $\color{#35bf28}+3.79\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7937ms 0.4435ms 2.2549 KOps/s 2.0438 KOps/s $\textbf{\color{#35bf28}+10.33\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9191ms 5.6356ms 177.4440 Ops/s 177.3073 Ops/s $\color{#35bf28}+0.08\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1650ms 0.3009ms 3.3234 KOps/s 3.4094 KOps/s $\color{#d91a1a}-2.52\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6031ms 0.3084ms 3.2423 KOps/s 3.3020 KOps/s $\color{#d91a1a}-1.81\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0875ms 5.6478ms 177.0593 Ops/s 177.4573 Ops/s $\color{#d91a1a}-0.22\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0096ms 0.2814ms 3.5535 KOps/s 3.1259 KOps/s $\textbf{\color{#35bf28}+13.68\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4548ms 0.2608ms 3.8342 KOps/s 3.4217 KOps/s $\textbf{\color{#35bf28}+12.06\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0761ms 5.8444ms 171.1031 Ops/s 174.3844 Ops/s $\color{#d91a1a}-1.88\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.4142ms 0.4733ms 2.1126 KOps/s 2.2407 KOps/s $\textbf{\color{#d91a1a}-5.72\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8159ms 0.4630ms 2.1600 KOps/s 1.9586 KOps/s $\textbf{\color{#35bf28}+10.29\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.3117ms 4.8682ms 205.4162 Ops/s 201.5847 Ops/s $\color{#35bf28}+1.90\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.2791ms 2.1907ms 456.4696 Ops/s 428.8004 Ops/s $\textbf{\color{#35bf28}+6.45\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.1127ms 0.9183ms 1.0890 KOps/s 847.7339 Ops/s $\textbf{\color{#35bf28}+28.46\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5750s 16.4281ms 60.8714 Ops/s 49.3707 Ops/s $\textbf{\color{#35bf28}+23.29\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.1966ms 1.8843ms 530.6954 Ops/s 516.6863 Ops/s $\color{#35bf28}+2.71\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.0812ms 1.1856ms 843.4868 Ops/s 767.4044 Ops/s $\textbf{\color{#35bf28}+9.91\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 9.5782ms 5.2328ms 191.1022 Ops/s 191.9759 Ops/s $\color{#d91a1a}-0.46\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.2115ms 1.9852ms 503.7258 Ops/s 519.5157 Ops/s $\color{#d91a1a}-3.04\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.3071ms 1.1041ms 905.7508 Ops/s 920.4341 Ops/s $\color{#d91a1a}-1.60\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 39.6929ms 36.4704ms 27.4195 Ops/s 28.3526 Ops/s $\color{#d91a1a}-3.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.5253ms 17.8124ms 56.1406 Ops/s 56.7480 Ops/s $\color{#d91a1a}-1.07\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 41.5869ms 36.6697ms 27.2705 Ops/s 27.2023 Ops/s $\color{#35bf28}+0.25\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.2099ms 17.8890ms 55.9004 Ops/s 56.0306 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 39.9912ms 38.0047ms 26.3125 Ops/s 26.0265 Ops/s $\color{#35bf28}+1.10\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.8941ms 19.3109ms 51.7844 Ops/s 51.2710 Ops/s $\color{#35bf28}+1.00\%$
test_storage_write_lazystack[50-img_shape0-small] 0.8781ms 0.2125ms 4.7060 KOps/s 4.4958 KOps/s $\color{#35bf28}+4.68\%$
test_storage_write_lazystack[100-img_shape1-atari] 1.6826ms 1.3181ms 758.6855 Ops/s 720.4014 Ops/s $\textbf{\color{#35bf28}+5.31\%}$
test_storage_write_lazystack[100-img_shape2-large_img] 2.4068ms 2.2131ms 451.8565 Ops/s 443.9524 Ops/s $\color{#35bf28}+1.78\%$
test_storage_write_lazystack[200-img_shape3-large_batch] 3.0681ms 2.9022ms 344.5626 Ops/s 337.3449 Ops/s $\color{#35bf28}+2.14\%$
test_storage_write_contiguous[50-img_shape0-small] 0.3165ms 0.1580ms 6.3288 KOps/s 6.3769 KOps/s $\color{#d91a1a}-0.75\%$
test_storage_write_contiguous[100-img_shape1-atari] 0.3754ms 0.2262ms 4.4211 KOps/s 4.3930 KOps/s $\color{#35bf28}+0.64\%$
test_storage_write_contiguous[100-img_shape2-large_img] 2.0064ms 1.7920ms 558.0489 Ops/s 560.7460 Ops/s $\color{#d91a1a}-0.48\%$
test_storage_write_contiguous[200-img_shape3-large_batch] 1.4919ms 1.3427ms 744.7843 Ops/s 725.4401 Ops/s $\color{#35bf28}+2.67\%$
test_collector_stack_then_write[50-img_shape0-small] 1.6032ms 1.1229ms 890.5770 Ops/s 894.1689 Ops/s $\color{#d91a1a}-0.40\%$
test_collector_stack_then_write[100-img_shape1-atari] 3.7334ms 3.5665ms 280.3867 Ops/s 277.7843 Ops/s $\color{#35bf28}+0.94\%$
test_collector_stack_then_write[100-img_shape2-large_img] 11.1592ms 5.6794ms 176.0758 Ops/s 177.4507 Ops/s $\color{#d91a1a}-0.77\%$
test_collector_stack_then_write[200-img_shape3-large_batch] 7.4241ms 7.0005ms 142.8464 Ops/s 140.9761 Ops/s $\color{#35bf28}+1.33\%$
test_collector_lazystack_then_write[50-img_shape0-small] 0.4319ms 0.2758ms 3.6253 KOps/s 3.6452 KOps/s $\color{#d91a1a}-0.54\%$
test_collector_lazystack_then_write[100-img_shape1-atari] 1.7087ms 1.4596ms 685.1175 Ops/s 650.5546 Ops/s $\textbf{\color{#35bf28}+5.31\%}$
test_collector_lazystack_then_write[100-img_shape2-large_img] 2.4497ms 2.2952ms 435.6944 Ops/s 414.4987 Ops/s $\textbf{\color{#35bf28}+5.11\%}$
test_collector_lazystack_then_write[200-img_shape3-large_batch] 3.3348ms 3.1096ms 321.5894 Ops/s 314.6494 Ops/s $\color{#35bf28}+2.21\%$
test_collector_without_rb[100-img_shape0-atari] 33.1316ms 32.3418ms 30.9198 Ops/s 30.6048 Ops/s $\color{#35bf28}+1.03\%$
test_collector_without_rb[200-img_shape1-large_batch] 0.6192s 99.1757ms 10.0831 Ops/s 15.5892 Ops/s $\textbf{\color{#d91a1a}-35.32\%}$
test_collector_with_rb[100-img_shape0-atari] 37.8198ms 36.9720ms 27.0475 Ops/s 27.1181 Ops/s $\color{#d91a1a}-0.26\%$
test_collector_with_rb[200-img_shape1-large_batch] 73.6691ms 72.6746ms 13.7600 Ops/s 13.8487 Ops/s $\color{#d91a1a}-0.64\%$
test_collector_without_rb_cuda[100-img_shape0-atari] 57.2191ms 55.2998ms 18.0833 Ops/s 18.1737 Ops/s $\color{#d91a1a}-0.50\%$
test_collector_without_rb_cuda[200-img_shape1-large_batch] 0.1105s 0.1100s 9.0933 Ops/s 9.0709 Ops/s $\color{#35bf28}+0.25\%$
test_collector_with_rb_cuda[100-img_shape0-atari] 59.4383ms 57.5791ms 17.3674 Ops/s 17.4827 Ops/s $\color{#d91a1a}-0.66\%$
test_collector_with_rb_cuda[200-img_shape1-large_batch] 0.1149s 0.1140s 8.7694 Ops/s 8.7893 Ops/s $\color{#d91a1a}-0.23\%$

vmoens added a commit that referenced this pull request Feb 16, 2026
dm-control 1.0.37 (released Feb 13) adds support for MuJoCo 3.5.0.
The <3.5 pin from #3506 now causes CI failures: uv resolves dm-control
to 1.0.37 (pulling mujoco 3.5.0), then the explicit pin downgrades
mujoco to 3.4.0, breaking dm-control which references the new
actuator_delay attribute added in 3.5.0.

Remove the upper bound across all CI scripts, workflows, and docs.

Co-authored-by: Cursor <cursoragent@cursor.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CI Has to do with CI setup (e.g. wheels & builds, tests...) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Documentation Improvements or additions to documentation

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant