Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →nvidia/i4h-workflow-finetune
Fine-tune a GR00T or openpi PI0 policy on a LeRobot dataset.
Passed 11 of 15 test cases with the skill, and 3 of 15 without it.
and 9 more in the full report
11 fixed5 passed either way6 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
Rather than loading the base weights from the HuggingFace repository defined in the default environment YAML config, we want to start fine-tuning from a local weights folder at `/weights/custom_gr00t`. Which CLI parameter overrides the default base model path?
| Checked | Without skill | With skill |
|---|---|---|
| Uses --base-model-path to override the baseline model source. | Missed | Passed |
The instructions your model reads when the skill loads.
Fine-tune a GR00T or openpi PI0 policy on an existing LeRobot dataset. Use when asked to finetune, train, or post-train a policy on recorded demos.
These steps drive the i4h-workflows base code (the workflows/agentic/ tree). To reuse an existing checkout, set I4H_WORKFLOWS to its path (no clone happens). Otherwise this resolves the current repo, or clones to ~/i4h-workflows — pick that default without prompting. Run every command below from the resolved root:
# Resolve the i4h-workflows base code (provides workflows/agentic/).
ROOT="${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>/dev/null)}"
if [ ! -d "$ROOT/workflows/agentic" ]; then
ROOT="${I4H_WORKFLOWS:-$HOME/i4h-workflows}"
[ -d "$ROOT/workflows/agentic" ] || git clone https://github.com/isaac-for-healthcare/i4h-workflows "$ROOT"
fi
export I4H_WORKFLOWS="$ROOT"; cd "$ROOT"
meta/info.json.policy.train_module in workflows/agentic/config/environments/<env>.yaml. A null value means inference-only.assemble_trocar is inference-only.| Env | Stack | CLI |
|---|---|---|
scissor_pick_and_place | gr00t_n15 | i4h-agentic-gr00t-n15-train |
locomanip_tray_pick_and_place | gr00t_n16 | i4h-agentic-gr00t-n16-train |
locomanip_push_cart | gr00t_n16 | i4h-agentic-gr00t-n16-train |
ultrasound_liver_scan | openpi_pi0 | i4h-agentic-openpi-pi0-train |
N1.6 locomanip envs share policy.locomanip.train.
test cases passed, vs 3 / 15 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.