Downloads · 30 days
0
RLobot-jun/gr00t-n17-bigenlight-50per-task-svf-iql-k04-g025-step5000
gr00t-n17-bigenlight-50per-task-svf-iql-k04-g025-step5000 is a machine learning model from RLobot-jun. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Lightweight inference export; BC weights are NOT included.
Downloads · 30 days
0
Access
Public
Updated Sep 21, 2026
Repo size
100 MB
Likes
0
Public
Click a slice to open those files.
.safetensors100 MB · 96%
From the Hugging Face model README
Lightweight inference export; BC weights are NOT included.
RLobot-jun/gr00t-n17-bigenlight-50per-task-step10000, pinned revision 1704897ac6a2a93c1d1fd806925b9237977ed8c5.adapter/actor_lora.safetensors: actor adapters only.adapter/env_q.safetensors: frozen online Q1/Q2.adapter/inner_critic.safetensors: trained online inner critics.adapter/adapter_config.json.Load the pinned BC and its processor first, then apply adapters to a matching LoRA-enabled ProjectedFlowActor. Do not apply on top of an already SVF-finetuned actor. See adapter/README.md. Root processor/statistics files are copies from the base BC. No optimizer, reference, target networks or VLM weights included. Not a native standalone BC checkpoint or training-resume bundle. Frozen actor tensors were checked against the local BC after training dtype conversion; full robot inference parity was not tested for this export.