4D-RGPT-8B
nvidia/4D-RGPT-8B
- kind
- original
- quant
- bf16 (16-bit)
- family
- misc
- modality
- video-text-to-text
- root model
- Efficient-Large-Model/NVILA-Lite-8B
- provider
- nvidia
- license
- cc-by-nc-4.0
- language
- en
- context
- —
- library
- transformers
- vRAM (GB)
- —
- size on disk
- —
- downloads
- 103
- likes
- 17
- created
- —
- last modified
- 2026-06-02T05:03:07+00:00 (2026)
- huggingface
- https://huggingface.co/nvidia/4D-RGPT-8B
- tags
- transformerssafetensorsllava_llamamultimodalvideo-understandingregion-grounding3d-reasoning4d-reasoningperceptual-distillationnvilavilavideo-text-to-textendataset:nvidia/R4D-Bencharxiv:2512.17012base_model:Efficient-Large-Model/NVILA-Lite-8Bbase_model:finetune:Efficient-Large-Model/NVILA-Lite-8Blicense:cc-by-nc-4.0endpoints_compatibleregion:us
Benchmark scores (0)
no scores extracted for this variant.
Discussions (0)
none scraped.