<div id="top" align="center">
<p align="center">
<img src="https://ik.imagekit.io/StarBurger/SimScale/title_1080p.gif">
</p>
Learning to Drive via Real-World Simulation at Scale
</div>
<div id="top" align="center">
<p align="center">
<img src="assets/teaser.png" >
</p>
</div>
Haochen Tian,
Tianyu Li,
Haochen Liu,
Jiazhi Yang,
Yihang Qiu,
Guang Li,
Junli Wang,
Yinfeng Gao,
Zhang Zhang,
Liang Wang,
Hangjun Ye,
Tieniu Tan,
Long Chen,
Hongyang Li
- 📧 Primary Contact: Haochen Tian (tianhaochen2023@ia.ac.cn)
- 📜 Materials:
🌐 𝕏 |
📰 Media |
🗂️ Slides |
🪧 Poster |
🎬 Talk (in Chinese)
- 🖊️ Joint effort by CASIA, OpenDriveLab at HKU, and Xiaomi EV.
🔥 Highlights
- 🏗️ A scalable simulation pipepline that synthesizes diverse and high-fidelity reactive driving scenarios with pseudo-expert demonstrations.
- 🚀 An effective sim-real co-training strategy that improves robustness and generalization synergistically across various end-to-end planners.
- 🔬 A comprehensive recipe that reveals crucial insights into the underlying scaling properties of sim-real learning systems for end-to-end autonomy.
📢 News
[2025/2/21] 🎉 Accepted to CVPR 2026.[2025/1/16] We released the data and models on 👾 ModelScope to better serve users in China.[2026/1/6] We released the code v1.0.[2025/12/31] We released the data and models v1.0 on 🤗 Hugging Face. Happy New Year ! 🎄[2025/12/1] We released our paper on arXiv.
📋 TODO List
- [x] More Visualization Results.
- [x] Future Sensors Data.
- [x] Sim-Real Co-training Code release (Jan. 2026).
- [x] Simulation Data release (Dec. 2025).
- [x] Checkpoints release (Dec. 2025).
📌 Table of Contents
🏛️ Model Zoo
<table>
<tr style="text-align: center;">
<th rowspan="2">Model</th>
<th rowspan="2">Backbone</th>
<th rowspan="2">Sim-Real Config</th>
<th colspan="2">NAVSIM v2 navhard</th>
<th colspan="2">NAVSIM v2 navtest</t>
</tr>
<tr style="text-align: center;">
<th>EPDMS</th>
<th>CKPT</th>
<th>EPDMS</th>
<th>CKPT</th>
</tr>
<!-- LTF -->
<tr>
<td><a href="./navsim/planning/script/config/common/agent/transfuser_agent.yaml">LTF</a></td>
<td>ResNet34</td>
<td>w/ pseudo-expert</td>
<td><a href="./assets/csv/LTF/ltf_sim_navhard.csv">30.3</a> | +6.9</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/LTF/ltf_sim_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FLTF%2Fltf_sim_navtest.ckpt?id=170866">MS</a></td>
<td><a href="./assets/csv/LTF/ltf_sim_navtest.csv">84.4</a> | +2.9</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/LTF/ltf_sim_navtest.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FLTF%2Fltf_sim_navhard.ckpt?id=170866">MS</a></td>
</td>
</tr>
<!-- DiffusionDrive -->
<tr>
<td><a href="./navsim/planning/script/config/common/agent/diffusiondrive_agent.yaml">DiffusionDrive</a></td>
<td>ResNet34</td>
<td>w/ pseudo-expert</td>
<td><a href="./assets/csv/DiffusionDrive/diffusiondrive_sim_navhard.csv">32.6</a> | +5.1</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/DiffusionDrive/diffusiondrive_sim_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FDiffusionDrive%2Fdiffusiondrive_sim_navhard.ckpt?id=170866">MS</a></td>
<td><a href="./assets/csv/DiffusionDrive/diffusiondrive_sim_navtest.csv">85.9</a> | +1.7</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/DiffusionDrive/diffusiondrive_sim_navtest.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FDiffusionDrive%2Fdiffusiondrive_sim_navtest.ckpt?id=170866">MS</a></td>
</tr>
<!-- GTRS-Dense block -->
<tr>
<td rowspan="4"><a href="./navsim/planning/script/config/common/agent/gtrs_dense_vov.yaml">GTRS-Dense</a></td>
<td rowspan="2">ResNet34</td>
<td>w/ pseudo-expert</td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_resnet_sim_expert_navhard.csv">46.1</a> | +7.8</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_resnet_sim_expert_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_resnet_sim_expert_navhard.ckpt?id=170866">MS</a></td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_resnet_sim_expert_navtest.csv">84.0</a> | +1.7</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_resnet_sim_expert_navtest.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_resnet_sim_expert_navtest.ckpt?id=170866">MS</a></td>
</tr>
<tr>
<td>rewards only</td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_resnet_sim_expert_navhard.csv">46.9</a> | +8.6</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_resnet_sim_reward_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_resnet_sim_reward_navhard.ckpt?id=170866">MS</a></td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_resnet_sim_reward_navtest.csv">84.6</a> | +2.3</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_resnet_sim_reward_navtest.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_resnet_sim_reward_navtest.ckpt?id=170866">MS</a></td>
</tr>
<tr>
<td rowspan="2">V2-99</td>
<td>w/ pseudo-expert</td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_vov_sim_expert_navhard.csv">47.7</a> | +5.8</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_vov_sim_expert_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_vov_sim_expert_navhard.ckpt?id=170866">MS</a></td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_vov_sim_expert_navtest.csv">84.5</a> | +0.5</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_vov_sim_expert_navtest.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense_vov_sim_expert_navtest.ckpt?id=170866">MS</a></td>
</tr>
<tr>
<td>rewards only</td>
<td><a href="./assets/csv/GTRS_Dense/gtrs_dense_vov_sim_reward_navhard.csv">48.0</a> | +6.1</td>
<td><a href="https://huggingface.co/datasets/OpenDriveLab/SimScale/blob/main/SimScale_ckpts/GTRS_Dense/gtrs_dense_vov_sim_reward_navhard.ckpt">HF</a> /
<a href="https://www.modelscope.cn/datasets/OpenDriveLab/SimScale/file/view/master/SimScale_ckpts%2FGTRS_Dense%2Fgtrs_dense
