Simulation & AI Compute InfraExplainer
How Much GPU Memory Does SmolVLA Fine-Tuning Need? A VRAM Calculation
Split SmolVLA fine-tuning memory into weights, gradients, optimizer state and activations, then size batches for 24 GB and 48 GB GPUs from …
LeRobot
- Physical AI Models & Core TechComparison
Is 50 Demos Enough to Fine-Tune a VLA? ACT vs Diffusion Policy vs SmolVLA
Demo counts, training setups and latency reported for ACT, Diffusion Policy and SmolVLA, and why pretraining lifted success on the same 50 …
- Simulation & AI Compute InfraTutorial
How to Install MuJoCo and LeRobot: Matching Versions by OS, With or Without a GPU
Set up MuJoCo 3.16.0 and LeRobot 0.6.1 in one Python 3.12 environment: per-OS wheel support, CUDA driver floors, headless rendering with …
- Simulation & AI Compute InfraTutorial
How to Fine-Tune Isaac GR00T N1.7: From LeRobot-Format Data to Open-Loop Evaluation
Step-by-step GR00T N1.7 fine-tuning from the official repo: install, LeRobot v2 data and modality.json, config, training, open-loop …
- Physical AI Models & Core TechTutorial
How to Train an ACT Policy on the SO-101 Arm with LeRobot: From Teleoperation Data to Rollout
Set up SO-101 leader and follower arms with Hugging Face LeRobot, record a teleoperation dataset, then train and run an ACT policy, with …