Team Ai
2 shown

datasets

Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.

Clear all
01arrow-hf /wam-landscape-and-deployment-20260907 WAM 选型地图与部署实测 · 2026-09 World-Action Model(世界-动作模型)在 2026 年 9 月的开源格局调研,六个候选在 NVIDIA RTX PRO 4000 Blackwell(sm_120) 上的真实部署实测, 以及这些模型的算子在 Ascend 910C(A3) 上的存在性与性能对照。 调研覆盖约 90 个模型与系统;部署部分是在单张 24 GB 消费级工作站卡上亲手跑出来的, 包含能力契约、逐层架构、算子剖析与性能数字。 为什么做这个 大多数 WAM 论文只报成功率,很少报能力契约(几路相机、动作是什么含义、世界模型那一半推理时还在不在), 几乎不报算子构成(移植到别的加速器要手写什么)。而跨论文的成功率数字大面积不可比。 这份材料试图补上这三块,并且把每个数字的口径写清楚。 五条主要结论 1. 「WAM 比 VLA 强」目前是假设,不是已验证结论。 唯一一篇第三方统一重跑的鲁棒性研究里,纯 VLA 的 π0.5 在最严的分布外评测上得分最高,… See the full description on the dataset page: https://huggingface.co/datasets/arrow-hf/wam-landscape-and-deployment-20260907.imagen<1K0 likes393 downloads1mo agoHugging Face02Eyas /reli3d-deployment-artifacts ReLi3D — Deployment Artifacts & Demo Reconstructions Artifacts from deploying ReLi3D (Relightable Multi-view 3D Reconstruction with Disentangled Illumination, Dihlmann et al., ICLR 2026) and running feed-forward inference on multi-view objects. Model: StabilityLabs/ReLi3D · Code: Stability-AI/ReLi3D · Paper: arXiv:2603.19753 Given a handful of posed RGBA views of an object, ReLi3D predicts in ~1 second a UV-unwrapped PBR mesh (baseColor + metallic-roughness + normal) and the… See the full description on the dataset page: https://huggingface.co/datasets/Eyas/reli3d-deployment-artifacts.imageimage-to-3dn<1K0 likes84 downloads4mo agoHugging Face

Listings come live from the Hugging Face Hub API. Team Ai does not host these files.