High-standard model card (FlashDrive); NVIDIA non-commercial license

README.md

@@ -1,82 +1,64 @@
 ---
-pipeline_tag: robotics
-library_name: transformers
 license: other
+license_name: nvidia-license
+license_link: https://huggingface.co/nvidia/Alpamayo-1.5-10B/blob/main/LICENSE
+pipeline_tag: robotics
 language:
 - en
+inference: false
 base_model:
 - z-lab/Alpamayo-R1-10B
 tags:
 - flashdrive
 - dflash
 - speculative-decoding
-- speculative-decoding-draft
 - block-diffusion
 - draft-model
 - autonomous-driving
 - vision-language-action
 ---
 
-# Alpamayo-R1-10B-DFlash
-
-[Paper](https://arxiv.org/abs/2602.06036) | [GitHub](https://github.com/z-lab/dflash) | [Blog](https://z-lab.ai/projects/dflash) | [FlashDrive](https://github.com/z-lab/flashdrive)
+# Alpamayo 1 (R1) — DFlash draft
 
-[DFlash](https://github.com/z-lab/dflash) draft model for
-[z-lab/Alpamayo-R1-10B](https://huggingface.co/z-lab/Alpamayo-R1-10B), used by
-[FlashDrive](https://github.com/z-lab/flashdrive) to accelerate the chain-of-causation
-reasoning stage.
+[Paper](https://arxiv.org/abs/2602.06036) | [DFlash](https://github.com/z-lab/dflash) | [GitHub](https://github.com/z-lab/flashdrive) | [Blog](https://z-lab.ai/projects/flashdrive/) | [Models](https://huggingface.co/collections/z-lab/flashdrive)
 
-DFlash uses a lightweight block-diffusion draft to propose multiple tokens in parallel;
-the target model verifies each block in a single forward, preserving its output
-distribution. This draft is a 2-layer Qwen3-style network (block size 8) conditioned on
-target hidden states captured from layers 24/30/31/32/34. The repository also ships
-`mask_embedding.pt`, the trained mask-token embedding that FlashDrive appends to the
-target's embedding table.
+[DFlash](https://github.com/z-lab/dflash) draft model for [z-lab/Alpamayo-R1-10B](https://huggingface.co/z-lab/Alpamayo-R1-10B), used by [FlashDrive](https://github.com/z-lab/flashdrive) to accelerate the chain-of-causation reasoning of [Alpamayo 1 (R1)](https://huggingface.co/nvidia/Alpamayo-R1-10B).
 
-**This is not a standalone language model.** FlashDrive attaches it to the base checkpoint
-automatically, derived from the base repo id by the `-DFlash` suffix.
+DFlash (ICML 2026) uses a lightweight block-diffusion draft to propose several tokens in parallel; the target verifies each block in a single forward, preserving its output distribution. This draft is a 2-layer Qwen3-style network (block size 8) conditioned on target hidden states from layers 24/30/31/32/34. The repository also ships `mask_embedding.pt`, the trained mask-token embedding FlashDrive appends to the target's embedding table.
 
-## Quick Start
+> [!NOTE]
+> **Not a standalone language model.** FlashDrive attaches it to the [base checkpoint](https://huggingface.co/z-lab/Alpamayo-R1-10B) automatically — you do not load this repository directly.
 
-Install [FlashDrive](https://github.com/z-lab/flashdrive), then load the base checkpoint —
-the `-PARO` and `-DFlash` companions are derived from it by suffix and fetched automatically:
+## Usage
 
 ```python
-from flashdrive import load_flashdrive_model
+import flashdrive
 
-model = load_flashdrive_model("z-lab/Alpamayo-R1-10B")
-
-pred_xyz, pred_rot = model.sample_trajectories_streaming(data)
+# from_pretrained fetches this -DFlash checkpoint automatically
+model = flashdrive.from_pretrained("z-lab/Alpamayo-R1-10B")
 ```
 
-The first call per stream only prefills the KV cache and returns `(None, None)`.
-For an end-to-end benchmark on a PhysicalAI-AV clip:
-
-```bash
-python scripts/infer.py --model_path z-lab/Alpamayo-R1-10B
-```
+See the [base model card](https://huggingface.co/z-lab/Alpamayo-R1-10B) and the [FlashDrive repository](https://github.com/z-lab/flashdrive) for the full pipeline.
 
 ## License
 
-These weights are derived from NVIDIA's Alpamayo release and remain governed by its
-[non-commercial license](https://huggingface.co/nvidia/Alpamayo-R1-10B/blob/main/LICENSE).
-The [FlashDrive](https://github.com/z-lab/flashdrive) inference code is MIT.
+This checkpoint is derived from NVIDIA's Alpamayo weights and is governed by the [NVIDIA License](https://huggingface.co/nvidia/Alpamayo-1.5-10B/blob/main/LICENSE), which permits **non-commercial use only** and extends to derivative works. The [FlashDrive](https://github.com/z-lab/flashdrive) inference code is separately released under the [MIT License](https://github.com/z-lab/flashdrive/blob/main/LICENSE).
 
 ## Citation
 
 ```bibtex
-@article{chen2026dflash,
-  title   = {{DFlash: Block Diffusion for Flash Speculative Decoding}},
-  author  = {Chen, Jian and Liang, Yesheng and Liu, Zhijian},
-  journal = {arXiv preprint arXiv:2602.06036},
-  year    = {2026}
+@inproceedings{chen2026dflash,
+  title     = {{DFlash: Block Diffusion for Flash Speculative Decoding}},
+  author    = {Chen, Jian and Liang, Yesheng and Liu, Zhijian},
+  booktitle = {International Conference on Machine Learning (ICML)},
+  year      = {2026}
 }
 ```
 
 ```bibtex
 @article{li2026flashdrive,
-      title={FlashDrive: Flash Vision-Language-Action Inference For Autonomous Driving},
-      author={Zekai Li, Yihao Liang, Hongfei Zhang, Jian Chen, Yesheng Liang, Zhijian Liu},
-      year={2026}
+  title  = {{FlashDrive: Flash Vision-Language-Action Inference for Autonomous Driving}},
+  author = {Li, Zekai and Liang, Yihao and Zhang, Hongfei and Chen, Jian and Liang, Yesheng and Liu, Zhijian},
+  year   = {2026}
 }
 ```