Search results

Allow policies to be instantiated without having the dataset they were trained on.

datasetIssues regarding data inputs, processing, or datasetsIssues regarding data inputs, processing, or datasets

enhancementSuggestions for new features or improvementsSuggestions for new features or improvements

policiesItems related to robot policiesItems related to robot policies

trainingIssues related at training timeIssues related at training time

Status: Open.

#3318 In huggingface/lerobot;

· nhnifong opened on Apr 8, 2026

Inference with ACT with relative action not working well

bugSomething isn’t working correctlySomething isn’t working correctly

documentationImprovements or fixes to the project’s docsImprovements or fixes to the project’s docs

enhancementSuggestions for new features or improvementsSuggestions for new features or improvements

policiesItems related to robot policiesItems related to robot policies

processorIssue related to processorIssue related to processor

Status: Open.

#3312 In huggingface/lerobot;

· LaFeuilleMorte opened on Apr 8, 2026

[TxRxResult] There is no status packet

bugSomething isn’t working correctlySomething isn’t working correctly

configurationProblems with configuration files or settingsProblems with configuration files or settings

Status: Open.

#3299 In huggingface/lerobot;

· nanyi98 opened on Apr 7, 2026

[Feature Request] LeRobot + Gemma + Unitree G1: Integrating Gemma as a VLA model

bugSomething isn’t working correctlySomething isn’t working correctly

documentationImprovements or fixes to the project’s docsImprovements or fixes to the project’s docs

enhancementSuggestions for new features or improvementsSuggestions for new features or improvements

policiesItems related to robot policiesItems related to robot policies

testsProblems with test coverage, failures, or improvements to testingProblems with test coverage, failures, or improvements to testing

trainingIssues related at training timeIssues related at training time

Status: Open.

#3298 In huggingface/lerobot;

· s-tory opened on Apr 6, 2026

🌐 [i18n-zh] Translating docs to Chinese

datasetIssues regarding data inputs, processing, or datasetsIssues regarding data inputs, processing, or datasets

documentationImprovements or fixes to the project’s docsImprovements or fixes to the project’s docs

evaluationFor issues or PRs related to environment evaluation, and benchmarks.For issues or PRs related to environment evaluation, and benchmarks.

examplesIssues related to the examplesIssues related to the examples

performanceIssues aimed at improving speed or resource usageIssues aimed at improving speed or resource usage

policiesItems related to robot policiesItems related to robot policies

processorIssue related to processorIssue related to processor

questionRequests for clarification or additional informationRequests for clarification or additional information

sensorsEverything related to sensorsEverything related to sensors

simulationMatters involving system simulation or modelingMatters involving system simulation or modeling

trainingIssues related at training timeIssues related at training time

Status: Open.

#3290 In huggingface/lerobot;

· tc-huang opened on Apr 6, 2026

Inquiry about Training Configurations (Batch Size, LR, Commit Hash) for Replicating SmolVLA on LIBERO Benchmark

evaluationFor issues or PRs related to environment evaluation, and benchmarks.For issues or PRs related to environment evaluation, and benchmarks.

performanceIssues aimed at improving speed or resource usageIssues aimed at improving speed or resource usage

policiesItems related to robot policiesItems related to robot policies

trainingIssues related at training timeIssues related at training time

Status: Open.

#3287 In huggingface/lerobot;

· Yoonkyo opened on Apr 5, 2026

RelativeActionsProcessorStep: should use action_names[0] when isinstance(self.action_names[0], list)=true

bugSomething isn’t working correctlySomething isn’t working correctly

datasetIssues regarding data inputs, processing, or datasetsIssues regarding data inputs, processing, or datasets

policiesItems related to robot policiesItems related to robot policies

Status: Open.

#3283 In huggingface/lerobot;

· xianglunkai opened on Apr 5, 2026

Dataset quality diagnostic tool -- catches silent training failures

CIIssues related to the continuous integration pipelineIssues related to the continuous integration pipeline

datasetIssues regarding data inputs, processing, or datasetsIssues regarding data inputs, processing, or datasets

dependenciesConcerns about external packages, libraries, or versioningConcerns about external packages, libraries, or versioning

trainingIssues related at training timeIssues related at training time

Status: Open.

#3280 In huggingface/lerobot;

· jashshah999 opened on Apr 3, 2026

Can some frontier world models be integrated?

enhancementSuggestions for new features or improvementsSuggestions for new features or improvements

Status: Open.

#3271 In huggingface/lerobot;

· moshilangzi opened on Apr 3, 2026

smolvla can not reproduce results in libero eval using official checkpoints

bugSomething isn’t working correctlySomething isn’t working correctly

datasetIssues regarding data inputs, processing, or datasetsIssues regarding data inputs, processing, or datasets

evaluationFor issues or PRs related to environment evaluation, and benchmarks.For issues or PRs related to environment evaluation, and benchmarks.

policiesItems related to robot policiesItems related to robot policies

simulationMatters involving system simulation or modelingMatters involving system simulation or modeling

Status: Open.

#3264 In huggingface/lerobot;

· zouyuanpeng opened on Apr 2, 2026

Question: Why does lerobot keep the entire vision_tower in float32, while the original openpi only preserves 3 specific embedding weights?

bugSomething isn’t working correctlySomething isn’t working correctly

policiesItems related to robot policiesItems related to robot policies

questionRequests for clarification or additional informationRequests for clarification or additional information

trainingIssues related at training timeIssues related at training time

Status: Open.

#3259 In huggingface/lerobot;

· XueSongTap opened on Apr 1, 2026

Cannot reproduce the performance of PI0 on LIBERO benchmark

bugSomething isn’t working correctlySomething isn’t working correctly

evaluationFor issues or PRs related to environment evaluation, and benchmarks.For issues or PRs related to environment evaluation, and benchmarks.

performanceIssues aimed at improving speed or resource usageIssues aimed at improving speed or resource usage

policiesItems related to robot policiesItems related to robot policies

Status: Open.

#3252 In huggingface/lerobot;

· jiuyixu25 opened on Mar 31, 2026