← Back to the library

Model · Task-specific robot configuration required

SmolVLA base checkpoint

LeRobot publishes a SmolVLA base model intended for fine-tuning on a specific robot task.

smolvlabase-modelfine-tuning
SmolVLA project artwork
SmolVLA project artworkHugging Face

About this project

The model card provides weights and examples for model loading, training and evaluation. Its stated role is a base for further fine-tuning. No ready-to-run SO101 task or success rate is asserted by this shared entry.

Setup & reproduction

Read the model card and install the matching LeRobot SmolVLA dependencies. Supply task demonstrations and verify the camera, state, action and processor configuration before evaluating a fine-tuned policy.

Sources checked 2026-09-21. Shared by a community member; original work by Hugging Face / LeRobot.

  • Source: https://huggingface.co/lerobot/smolvla_base Known limits:
  • Immutable resource revisions have not been captured; linked default branches/cards can change.
  • No independent installation, policy evaluation or hardware reproduction was performed.
  • SO101 task compatibility, trained task, camera mapping and task-specific evaluation are not established.

Restored from the original local preview catalog; original creator attribution and reviewed source notes are preserved.

Interactive preview

No interactive preview is available for this project. Explore the linked resources and demonstration above.

Discussion

Share what you tried, ask a question, or add a useful detail. Comments appear newest first.

Sign in to comment

Loading comments…

Sign in to report project