Release v0.6.1

⚠️ Breaking changes (v0.6.0 -> v0.6.1)

• lerobot.types module was renamed to lerobot.lerobot_types: #4232

What's Changed

• chore(dependencies): Bump lerobot to 0.6.1 by @imstevenpmwork in #3957
• Make doc builds faster by @mishig25 in #3958
• docs(GR00T): Update GR00T 1.7 LIBERO checkpoints by @liorbenhorin in #3961
• feat(annotate): improve VLM subtask annotation (legible contact sheets, seeded relabeling, self-hosted vLLM recipe) by @pkooij in #3896
• Fix readme when using annotations by @NikodemBartnik in #4029
• refactor(evo1): use transformers flash attention probe by @imstevenpmwork in #4013
• refactor(hub): load safetensors directly on target device by @imstevenpmwork in #4012
• Aggregate policy sub-losses through MetricsTracker by @Maximellerbach in #4024
• Fix annotation example by @NikodemBartnik in #4041
• Move annotation dependencies to module scope by @pkooij in #4040
• fix(safetensors): expand bare "cuda" to current device for safetensors loads by @Maximellerbach in #4042
• refactor(policies): resolve policy components by convention by @imstevenpmwork in #4015
• refactor(processors): share policy pipeline builders by @imstevenpmwork in #4016
• refactor(policies): extract shared model components by @imstevenpmwork in #4054
• refactor(wall-x): subclass native Transformers Qwen2.5-VL instead of vendoring it by @imstevenpmwork in #4035
• refactor(pi05): use shared VLA components by @imstevenpmwork in #4063
• refactor(pi0): use shared VLA components by @imstevenpmwork in #4062
• refactor(eo1): reuse shared VLA components by @imstevenpmwork in #4061
• refactor(pi0_fast): reuse shared VLA components by @imstevenpmwork in #4055
• refactor(xvla): subclass native Transformers Florence2 instead of vendoring it by @imstevenpmwork in #4089
• chore(ci): upgrade claude workflow by @imstevenpmwork in #4096
• chore(docs): update security policy to adopt HF standards by @imstevenpmwork in #4098
• refactor(smolvla): reuse shared VLA components by @imstevenpmwork in #4064
• feat(annotate): run lerobot-annotate on HF Jobs via --job.target by @pkooij in #4095
• refactor(pi05): remove unused variables in embed_suffix method by @dudududukim in #3263
• Feature omx docs by @Eunsung-kespion in #3421
• docs(readme): fix missing periods at end of sentences by @ykdojo in #3473
• fix(scripts): register third-party plugins in lerobot_setup_motors by @imstevenpmwork in #4123
• fix(gamepad): Gamepad on macos often does not need fallback by @imstevenpmwork in #4125
• feat(diffusion): add gradient checkpointing for memory optimization by @imstevenpmwork in #4127
• fix(openarms): stop set_zero_position()ing on connect by @nepyope in #4058
• fix(robots): add retries while recording motor ranges by @imstevenpmwork in #4126
• fix(envs): set LiberoEnvConfig.fps default to 20 to match robosuite by @imstevenpmwork in #4124
• feat(dataset): add slice support to LeRobotDataset.getitem by @imstevenpmwork in #4129
• chore(agents): adding additional infos to AGENTS.md and bring-your-own-policies.mdx by @Maximellerbach in #3904
• feat(robot): Make SO follower P coefficient configurable by @imstevenpmwork in #4142
• chore(dependencies): update uv.lock by @imstevenpmwork in #3963
• chore(deps): bump diffusers to >=0.38.0,<0.40.0 by @imstevenpmwork in #4145
• (depth image processing): excluding depth frames from the RGB to BGR image processing by @CarolinePascal in #4135
• feat(dataset): accept token argument for private HF Hub datasets by @imstevenpmwork in #4136
• fix conversion error when tag is not exists. by @k1000dai in #3656
• chore: enable Dependabot weekly GitHub Actions bumps by @hf-dependantbot-rollout[bot] in #3677
• fix(datasets): clear video frame staging on episode reset by @MihaiAnca13 in #3683
• fix(scripts): restore policy training mode after eval_policy() in lerobot-eval by @imstevenpmwork in #4162
• fix(record): connect teleoperator before robot to avoid watchdog jump by @imstevenpmwork in #4166
• fix(processor): avoid Hub fallback for missing local processor state by @WOLIKIMCHENG in #3715
• fix(rollout): account for policy revision loading by @imstevenpmwork in #4161
• fix(envs): make MetaWorld evaluation reproducible with seeded object positions by @landegt in #3727
• (fix) Add LeKiwi support to lerobot-replay CLI by @zuoxingdong in #3739
• feat(config): add multiprocessing option to DataLoader context and sets spawn as default by @imstevenpmwork in #4139
• fix(processors): wrong feature key dropped in delta-action transform_features by @imstevenpmwork in #4165
• feat(rl): add multiprocessing option to training pipeline and sets spawn as default + guard by @imstevenpmwork in #4140
• feat(lekiwi): support LeKiwi in the rollout/eval CLI by @zuoxingdong in #3742
• fix(cameras): release device handle when connect() setup fails by @imstevenpmwork in #4187
• feat(camera): add manual exposure, gain, and white balance options for RealSense cameras by @imstevenpmwork in #4130
• feat(record): add --dataset.no_stamp to opt out of repo_id timestamping by @imstevenpmwork in #4190
• feat(motors): add XH540-W150, XC330-T288, XC330-T181 to Dynamixel tables by @charlie8612 in #3815
• refactor(logging): standardize logging with getLogger(name) in scripts by @imstevenpmwork in #4192
• fix(dependencies): protect peft import by @imstevenpmwork in #4188
• fix(policy): honor revisions when loading PEFT checkpoints by @s1lent4gnt in #4189
• Handle resuming empty local datasets by @AlexandreEDMOND in #3859
• chore(mypy): cover annotations and transforms by @AlexandreEDMOND in #3860
• fix(utils): handle missing/unresponsive TTS on Linux by @imstevenpmwork in #4199
• fix(env): eval env lifecycle by @imstevenpmwork in #4194
• fix peft factory test mocking by @s1lent4gnt in #4201
• fix(eval): prevent eval_policy crash when start_seed is None and num_envs>1 by @imstevenpmwork in #4203
• fix(datasets): bound memory of augment_dataset_quantile_stats by sampling frames by @pepa99 in #3749
• perf(docker): split dependency install from source copy for CI layer caching by @imstevenpmwork in #4208
• fix(robots): retry SO follower/leader bus reads on transient Feetech errors by @imstevenpmwork in #4207
• docs: fix dataset split fraction example by @saime428 in #3936
• docs: fix wording in backward compatibility guide by @Old-Ding in #3938
• docs: fix wording in guides by @Old-Ding in #3939
• chore(datasets): add typing to aggregate helpers by @imstevenpmwork in #4211
• fix(evo1): re-pad normalizer stats when loading from checkpoint by @nepyope in #3945
• fix(robot): type FK-to-EE action features as ACTION not STATE by @imstevenpmwork in #4213
• docs(agent-guide): prioritize uv over pip in §4.1 install block by @Kunal-Somani in #3799
• fix(find-cameras): enforce sequential lifecycle and add configurable warmup by @AnesBenmerzoug in #3593
• fix(utils): reject zero-norm / invalid quaternions in Rotation by @Bartok9 in #3988
• fix(utils): raise ValueError from get_safe_torch_device by @Bartok9 in #3992
• fix(utils): preserve exc_info/stack_info in init_logging formatter by @imstevenpmwork in #4215
• fix(utils): mark Transition.complementary_info NotRequired by @imstevenpmwork in #4216
• fix(utils): allow any JSON payload in write_json - #3993 by @imstevenpmwork in #4217
• fix(utils): validate precise_sleep spin/margin args by @imstevenpmwork in #4218
• docs: fix env processor code fences and minor doc errors by @tc-huang in #3953
• fix assembly instructions typo - SO-ARM101 by @NikodemBartnik in #4008
• fix(utils): add MPS branch to torch RNG state serialization by @sunnydave234 in #4014
• fix(smolvla): freeze the intended VLM layers when train_expert_only=False by @zuoxingdong in #4019
• fix(rollout): reject incompatible RTC policies by @imstevenpmwork in #4228
• feat(teleoperators): add DAgger/HIL smooth handover support for BiSOLeader by @Mr-C4T in #4028
• chore(deps): bump draccus by @ntjohnson1 in #4033
• fix(train): keep checkpoint processor stats on resume by @jashshah999 in #4022
• fix(config): accept pretrained_model dir for --config_path on resume by @jashshah999 in #4023
• fix(envs): use RoboCasa task horizons by @anasm266 in #4037
• (fix) typo in xvla soft_prompt param key lookup by @nepyope in #4048
• refactor(types): change module name by @imstevenpmwork in #4232
• Fix Backtrackable.can_peek_back off-by-one contract violation by @Osamaali313 in #4065
• fix(datasets): stop frame errors being treated as shard exhaustion in StreamingLeRobotDataset by @imstevenpmwork in #4237
• feat(robot): mirror new configs from SO10X to its Bi manual counterpart by @imstevenpmwork in #4238
• fix(lekiwi): pin MJPG capture in the default camera config by @zuoxingdong in #4083
• fix(RGB only): remove the stereo module fallback when setting colors parameters on RealSense cameras by @CarolinePascal in #4225
• feat(lekiwi): stream observations as ZMQ multipart (-25% bandwidth) by @zuoxingdong in #4088
• test(datasets): guard batched-encoding video staging from post-save cleanup by @nevertmr in #4110
• fix(train): guard checkpoint saving against save_freq=0 by @ahmet-f-gumustas in #4112
• feat(dataset): Support streaming from HF Storage Buckets + Bump HF hub & datasets by @imstevenpmwork in #4236
• docs: fix broken and placeholder links by @youdie006 in #4097
• fix(policies): VLA-JEPA prepare model input to take state from index 0 by @Maximellerbach in #4243
• fix: Fix a spelling typo in the RTC evaluation docstring. by @nathon-lee in #4253
• fix(datasets): Fix metadata when parquet files columns have different orders by @brunomachado37 in #3964
• docs: fix RoboMME Docker build command by @cupkk in #4157
• feat(dataset tools): support task replacement mappings in modify_tasks by @CarolinePascal in #4244
• fix(datasets): support multi-dimensional per-frame features in add_features by @CarolinePascal in #4246
• chore(dependencies): update uv.lock by @imstevenpmwork in #4151
• fix(config): catch draccus DecodingError in CLI parsing - #4106 by @imstevenpmwork in #4260
• feat(transforms): add 8 robotics-relevant image augmentations by @imstevenpmwork in #4210
• fix(processor): diagnose legacy Hub checkpoints by @imstevenpmwork in #4261
• fix(datasets): improve get_safe_version raise by @imstevenpmwork in #4263
• fix(config): scope draccus --help output to already-resolved choices by @imstevenpmwork in #4265
• chore(deps): bump the actions group with 15 updates by @dependabot[bot] in #4164
• docs(libero): recommend lerobot/libero dataset, add reproducibility tips by @zuoxingdong in #4185
• fix(vla_jepa): use device-safe autocast instead of hardcoded bfloat16 by @Maximellerbach in #4158
• feat(rollout): add smooth_handover flag to episodic strategy config by @Maximellerbach in #4159
• feat(rollout): add smooth_handover flag to DAgger strategy config by @Maximellerbach in #4160
• docs(pi05): LIBERO quickstart with v3 dataset, update guidance by @zuoxingdong in #4186
• Improved main docs page by @NikodemBartnik in #4197
• fix(rollout): apply torch compile mode correctly by @imstevenpmwork in #4268
• chore(dependecies): bump setuptools + pytest + uv version in CI by @imstevenpmwork in #4270
• fix(edit-dataset): redirect dataset.root via backup_path from get_output_path by @CarolinePascal in #4271
• fix(so_follower): check the observation for None before copying it by @noron12234 in #4255
• feat(env): config to skip the discarded scene rebuild on reset Libero by @imstevenpmwork in #4272
• fix(libero): don't reset inside step() on termination by @imstevenpmwork in #4273
• fix(train): honor policy.use_amp when the policy has no dtype field by @imstevenpmwork in #4274
• feat(cameras): prefer MJPG when fourcc unspecified in ZMQ by @imstevenpmwork in #4277
• docs(evo1): add LIBERO reproduction recipe by @imstevenpmwork in #4278
• feat(ci): re-enable stale issues countdown by @imstevenpmwork in #4279
• perf(eval): stop simulating environments whose episode has ended by @dvdimitrov13 in #4247
• fix(lekiwi): retry LeKiwi bus reads on transient Feetech errors by @zuoxingdong in #4283
• fix(cameras): D405 RealSense connection timeout on startup by @Liyux3 in #3894
• fix broken link by @NikodemBartnik in #4304
• docs(third party): adding docs for noticeable third party packages by @CarolinePascal in #4026
• fix(video): only default to torchcodec when it can actually be loaded by @CarolinePascal in #4307
• fix(warmup): adding back the no-warmup path in the connect() method of the RealSense camera. by @CarolinePascal in #4301
• fix(aggregate stats): fix episode statistics aggregation by @CarolinePascal in #4276
• perf(pi0fast): stop FAST decoding at the end-of-action marker by @imstevenpmwork in #4275
• fix(sarm): fail fast on unusable episode annotations by @s1lent4gnt in #4306
• feat(train): enable buckets with streaming dataset by @imstevenpmwork in #4312
• chore(dependencies): update uv.lock by @imstevenpmwork in #4284

New Contributors

• @dudududukim made their first contribution in #3263
• @Eunsung-kespion made their first contribution in #3421
• @ykdojo made their first contribution in #3473
• @hf-dependantbot-rollout[bot] made their first contribution in #3677
• @MihaiAnca13 made their first contribution in #3683
• @WOLIKIMCHENG made their first contribution in #3715
• @landegt made their first contribution in #3727
• @charlie8612 made their first contribution in #3815
• @pepa99 made their first contribution in #3749
• @saime428 made their first contribution in #3936
• @Old-Ding made their first contribution in #3938
• @Kunal-Somani made their first contribution in #3799
• @AnesBenmerzoug made their first contribution in #3593
• @Bartok9 made their first contribution in #3988
• @sunnydave234 made their first contribution in #4014
• @Mr-C4T made their first contribution in #4028
• @ntjohnson1 made their first contribution in #4033
• @anasm266 made their first contribution in #4037
• @Osamaali313 made their first contribution in #4065
• @nevertmr made their first contribution in #4110
• @ahmet-f-gumustas made their first contribution in #4112
• @youdie006 made their first contribution in #4097
• @nathon-lee made their first contribution in #4253
• @cupkk made their first contribution in #4157
• @noron12234 made their first contribution in #4255
• @dvdimitrov13 made their first contribution in #4247
• @Liyux3 made their first contribution in #3894

Full Changelog: v0.6.0...v0.6.1