Downloads · 30 days
0
lygwm-review/LY-GWM-RoboCasa-Human300
LY-GWM-RoboCasa-Human300 is a machine learning model from lygwm-review. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as other.
The company-owned LY-GWM graph model, S38 reward scorer, and covered original code and weights are publicly available only for non-commercial research under the LY-GWM Non-Commercial Research License v1.0. Commercial…
Downloads · 30 days
0
Access
Public
Updated Sep 29, 2026
Repo size
7.6 GB
Likes
0
Public
Click a slice to open those files.
.safetensors7.6 GB · 99%
From the Hugging Face model README
The company-owned LY-GWM graph model, S38 reward scorer, and covered original code and weights are publicly available only for non-commercial research under the LY-GWM Non-Commercial Research License v1.0. Commercial use is prohibited, including commercial products, paid services and development directed toward commercial products or services. Research reproduction, research benchmarking, and academic publication are permitted subject to the license. See LICENSE_SCOPE.md for the precise component scope and exclusions.
GR00T and other third-party materials retain their original licenses. GR00T source-code licensing and model-weight licensing are separate. This is a source-available research release, not a blanket MIT/Apache-2.0 license or a grant of unrestricted commercial rights.
中文:公司自研且有权授权的 LY-GWM 代码与权重仅限非商业研究用途,禁止商用。 第三方组件继续适用各自许可证,具体范围见 LICENSE_SCOPE.md。
Public inference weights for the S42 four-candidate learned-reward evaluation. Code and technical report: https://github.com/SEIN-LYGWM/LY-GWM-RoboCasa-Human300-eval
S42: 207/2500 (8.28%), RoboCasa 1.0.1, split pretrain.
Atomic seen 150/900 (16.6667%); composite seen 38/800 (4.750%);
composite unseen 19/800 (2.375%).
| Path | Component |
|---|---|
gr00t/ | R5 fine-tuned GR00T-N1.5 checkpoint-120000; retained existing files |
lygwm/best.pt | W5 action-conditioned graph dynamics; retained existing file |
s42/scorer-best.pt | S38 binary state-reward head used for S42 selection |
s42/scorer-training-config.json | Original scorer configuration pinned by S42 inference |
GR00T generates four candidates; LY-GWM predicts their future features/state;
S38 scores them; the highest reward logit selects the action chunk. S38 has
no explicit goal input and is not a calibrated long-horizon success model.
Use reviewer_s42/ in the code repository; this is not an automatic
Transformers pipeline. Pin the HF revision containing the S42 additions when
reproducing. Tensor SHA256 values are in release_info.json.
The original cluster run completed; the new portable entrypoint has static/CPU checks only, without a completed GPU reproduction on a second machine. Compiled environments and simulator assets are separate dependencies. Full historical raw action traces are not included in the code publication subset.
Preserve existing component notices and weight terms. Public availability does
not create a replacement license; see MODEL_PROVENANCE.md. This update adds
inference artifacts, not a complete training reproduction release.