Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models
| Date | Stars |
|---|---|
| 2026-07-24 | 512 |
| 2026-07-25 | 513 |
| 2026-07-28 | 516 |
| 2026-07-30 | 516 |
| 2026-07-31 | 518 |
| 2026-08-06 | 518 |
Today
— stars today
This week
+2 stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.39%/day
# Cosmos-Drive-Dreams
### External Links: [Paper](https://github.com/nv-tlabs/Cosmos-Drive-Dreams/blob/main/assets/_arxiv__cosmos_drive_dream_paper.pdf) | [Arxiv Paper](https://arxiv.org/abs/2506.09042) | [Paper Website](https://research.nvidia.com/labs/toronto-ai/cosmos_drive_dreams)
### On This Page: [Models](#cosmos-drive-open-source-summary) | [Dataset](#cosmos-drive-dreams-dataset) | [Toolkits](#cosmos-drive-dreams-toolkits) | [SDG Pipeline](#cosmos-drive-dreams-sdg-pipeline)
This is the official code repository of Cosmos-Drive-Dreams - a Synthetic Data Generation (**SDG**) pipeline built on [Cosmos World Foundation Models](https://www.nvidia.com/en-us/ai/cosmos/) for generating diverse and challenging scenarios for Autonomous Vehicle use-cases.
We open-source our **model weights**, **pipeline toolkits**, and a **dataset** (including cosmos-generated videos, paired HDMap and LiDAR), which consists of **81,802** clips.
<p align="center">
<img src="assets/teaser.png" alt="Cosmos-Drive-Dream Teaser">
</p>
<video controls autoplay loop src=https://github.com/user-attachments/assets/af926ed2-6f93-4e9d-8afe-95095792e8d8></video>
## News
- 2025-10-29: **World Scenario Rendering** feature added to toolkits! This enhanced rendering mode (used in [Cosmos-Transfer 2.5](https://github.com/nvidia-cosmos/cosmos-transfer2.5)) provides high-fidelity 3D geometry-based control signals with rich laneline and bounding box patterns compared to traditional HDMap rendering. See the [toolkit](https://github.com/nv-tlabs/Cosmos-Drive-Dreams/blob/main/cosmos-drive-dreams-toolkits/README.md#world-scenario-rendering-new-in-cosmos-transfer25-2bautomultiview) for details.
- 2025-10-22: [Data preprocessing](cosmos-drive-dreams-toolkits/convert_lidar_pointcloud_to_rangemap.py), post-training and inference scripts of lidar tokenizer and diffuison [models](cosmos-transfer-lidargen/README.md) are released ! See [Huggingface](https://huggingface.co/collections/nvidia/cosmos-drive-dreams) for our model cards.
- 2025-06-10: Model, Toolkits, and Dataset (including cosmos-generated video, HDMap, and LiDAR) are released! Stay tuned for the paired GT RGB videos.
https://github.com/user-attachments/assets/43c5b921-ef23-4d5d-8ab4-2a7a58a0cb77
<p align="center">
Condition Videos: HDMap / LiDAR Depth / World Scenario
</p>
## Cosmos-Drive Open-source Summary
| **Name** | **Type** | **Link** |
|----------------------------------|------------|----------|
| **Cosmos-7B-AV-Sample** (Paper Sec. [2.1]) | model | [`base_model.pt`](https://huggingface.co/nvidia/Cosmos-Transfer1-7B-Sample-AV) |
| **Cosmos-7B-Multiview-AV-Sample** (Paper Sec. [2.1]) | model | [Huggingface Link](https://huggingface.co/nvidia/Cosmos-Predict1-7B-Text2World-Sample-AV-Multiview) |
| **Cosmos-Transfer1-7B-Sample-AV** (Paper Sec. [2.2]) | model | [Huggingface Link](https://huggingface.co/nvidia/Cosmos-Transfer1-7B-Sample-AV) |
| **Cosmos-7B-Single2Multiview-Sample-AV** (Paper Sec. [2.3]) | model | [Huggingface Link](https://huggingface.co/nvidia/Cosmos-Transfer1-7B-SingleToMultiView-Sample-AV) |
| **Cosmos-7B-LiDAR-GEN-Sample-AV** (Paper Sec. [3]) | model | [Huggingface link](https://huggingface.co/collections/nvidia/cosmos-drive-dreams) |
<!---
| **Cosmos-7B-Annotate-Sample-AV** (Paper Sec. [2.4]) | model | To be released soon |
| **Cosmos Toolkit** | SDG tool | [Cosmos Drive Dreams Toolkits](https://github.com/nv-tlabs/Cosmos-Drive-Dreams/tree/main/cosmos-drive-dreams-toolkits) |
| **RDS-HQ Dataset Subset** | dataset | [Huggingface Dataset](https://huggingface.co/datasets/nvidia/PhysicalAI-Autonomous-Vehicle-Cosmos-Drive-Dreams) |
| **Cosmos-Drive-Dreams Dataset** | dataset | [Huggingface Dataset](https://huggingface.co/datasets/nvidia/PhysicalAI-Autonomous-Vehicle-Cosmos-Drive-Dreams) |
-->
## Cosmos-Drive-Dreams Dataset
Cosmos-Drive-Dreams Dataset conExcerpt of 12,533 characters
Read on GitHub
Xuanchi Ren · University of Toronto | NVIDIA · Canada
25
13
9
8
huan ling · University of Toronto, Nvidia · Canada
4
Justin Ruan · Taiwan
1
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:5be9dd17d7593515, topic:video-generation
matched fp:5be9dd17d7593515, topic:autonomous-driving