chore(docs): Update README and add affordance download.

README.md

@@ -4,9 +4,9 @@
 [![📖 Documentation](https://img.shields.io/badge/📖-Documentation-blue)](https://horizonrobotics.github.io/EmbodiedGen/docs/)
 [![GitHub](https://img.shields.io/badge/GitHub-EmbodiedGen-black?logo=github)](https://github.com/HorizonRobotics/EmbodiedGen)
 [![📄 arXiv](https://img.shields.io/badge/📄-arXiv-b31b1b)](https://arxiv.org/abs/2506.10600)
-[![🎥 Video](https://img.shields.io/badge/🎥-Video-red)](https://www.youtube.com/watch?v=rG4odybuJRk)
+[![🎥 Video](https://img.shields.io/badge/🎥-Video-red)](https://youtu.be/swjDFEvjn14)
 [![🤗 Dataset](https://img.shields.io/badge/🤗-Dataset-blue)](https://huggingface.co/datasets/HorizonRobotics/EmbodiedGenData)
-[![中文介绍](https://img.shields.io/badge/中文介绍-07C160?logo=wechat&logoColor=white)](https://mp.weixin.qq.com/s/HH1cPBhK2xcDbyCK4BBTbw)
+<!-- [![中文介绍](https://img.shields.io/badge/中文介绍-07C160?logo=wechat&logoColor=white)](https://mp.weixin.qq.com/s/HH1cPBhK2xcDbyCK4BBTbw) -->
 [![🤗 Hugging Face](https://img.shields.io/badge/🤗-EmbodiedGen_Asset_Gallery-blue)](https://huggingface.co/spaces/HorizonRobotics/EmbodiedGen-Gallery-Explorer)
 [![🤗 Hugging Face](https://img.shields.io/badge/🤗-Image_to_3D_Demo-blue)](https://huggingface.co/spaces/HorizonRobotics/EmbodiedGen-Image-to-3D)
 [![🤗 Hugging Face](https://img.shields.io/badge/🤗-Text_to_3D_Demo-blue)](https://huggingface.co/spaces/HorizonRobotics/EmbodiedGen-Text-to-3D)
@@ -15,7 +15,7 @@
 > **From intent to executable 3D worlds.**
 > ***EmbodiedGen*** compiles language, images, and edit commands into **simulation-ready 3D worlds** — physically plausible assets, large-scale scenes, and task-driven interactive environments, deployable across every major robotics simulator.
 
-<img src="docs/landing/assets/img/overview.jpg" alt="EmbodiedGen V2 Overview" width="800"/>
+<img src="docs/landing/assets/img/overview.jpg" alt="EmbodiedGen V2 Overview" width="700"/>
 
 ---
 
@@ -273,8 +273,7 @@ If you use EmbodiedGen in your research or projects, please cite:
 ## 🙌 Acknowledgement
 
 EmbodiedGen builds upon the following amazing projects and models:
-🌟 [Trellis](https://github.com/microsoft/TRELLIS) | 🌟 [Hunyuan-Delight](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) | 🌟 [Segment Anything](https://github.com/facebookresearch/segment-anything) | 🌟 [Rembg](https://github.com/danielgatis/rembg) | 🌟 [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4) | 🌟 [Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) | 🌟 [Kolors](https://github.com/Kwai-Kolors/Kolors) | 🌟 [Aesthetic Score](http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html) | 🌟 [Pano2Room](https://github.com/TrickyGo/Pano2Room) | 🌟 [Diffusion360](https://github.com/ArcherFMY/SD-T2I-360PanoImage) | 🌟 [Kaolin](https://github.com/NVIDIAGameWorks/kaolin) | 🌟 [diffusers](https://github.com/huggingface/diffusers) | 🌟 [gsplat](https://github.com/nerfstudio-project/gsplat) | 🌟 [ChatGPT](https://platform.openai.com/docs/models/gpt-4o) | 🌟 [SD3.5](https://huggingface.co/stabilityai/stable-diffusion-3.5-medium) | 🌟 [ManiSkill](https://github.com/haosulab/ManiSkill) | 🌟 [SAM3D](https://github.com/facebookresearch/sam-3d-objects) | 🌟 [infinigen](https://github.com/princeton-vl/infinigen)
-🌟 [Trellis](https://github.com/microsoft/TRELLIS) | 🌟 [Hunyuan-Delight](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) | 🌟 [Hunyuan3D-Part](https://github.com/Tencent-Hunyuan/Hunyuan3D-Part) | 🌟 [GraspGen](https://github.com/NVlabs/GraspGen) | 🌟 [Segment Anything](https://github.com/facebookresearch/segment-anything) | 🌟 [Rembg](https://github.com/danielgatis/rembg) | 🌟 [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4) | 🌟 [Stable Diffusion x4](https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler) | 🌟 [Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) | 🌟 [Kolors](https://github.com/Kwai-Kolors/Kolors) | 🌟 [ChatGLM3](https://github.com/THUDM/ChatGLM3) | 🌟 [Aesthetic Score](http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html) | 🌟 [Pano2Room](https://github.com/TrickyGo/Pano2Room) | 🌟 [Diffusion360](https://github.com/ArcherFMY/SD-T2I-360PanoImage) | 🌟 [Kaolin](https://github.com/NVIDIAGameWorks/kaolin) | 🌟 [diffusers](https://github.com/huggingface/diffusers) | 🌟 [gsplat](https://github.com/nerfstudio-project/gsplat) | 🌟 [QWEN-2.5VL](https://github.com/QwenLM/Qwen2.5-VL) | 🌟 [GPT4o](https://platform.openai.com/docs/models/gpt-4o) | 🌟 [SD3.5](https://huggingface.co/stabilityai/stable-diffusion-3.5-medium) | 🌟 [ManiSkill](https://github.com/haosulab/ManiSkill) | 🌟 [SAM3D](https://github.com/facebookresearch/sam-3d-objects) | 🌟 [infinigen](https://github.com/princeton-vl/infinigen)
+🌟 [Trellis](https://github.com/microsoft/TRELLIS) | 🌟 [Hunyuan-Delight](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) | 🌟 [Hunyuan3D-Part](https://github.com/Tencent-Hunyuan/Hunyuan3D-Part) | 🌟 [GraspGen](https://github.com/NVlabs/GraspGen) | 🌟 [Segment Anything](https://github.com/facebookresearch/segment-anything) | 🌟 [Rembg](https://github.com/danielgatis/rembg) | 🌟 [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4) | 🌟 [Stable Diffusion x4](https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler) | 🌟 [Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) | 🌟 [Kolors](https://github.com/Kwai-Kolors/Kolors) | 🌟 [ChatGLM3](https://github.com/THUDM/ChatGLM3) | 🌟 [Aesthetic Score](http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html) | 🌟 [Pano2Room](https://github.com/TrickyGo/Pano2Room) | 🌟 [Diffusion360](https://github.com/ArcherFMY/SD-T2I-360PanoImage) | 🌟 [Kaolin](https://github.com/NVIDIAGameWorks/kaolin) | 🌟 [diffusers](https://github.com/huggingface/diffusers) | 🌟 [gsplat](https://github.com/nerfstudio-project/gsplat) | 🌟 [QWEN-2.5VL](https://github.com/QwenLM/Qwen2.5-VL) | 🌟 [SD3.5](https://huggingface.co/stabilityai/stable-diffusion-3.5-medium) | 🌟 [ManiSkill](https://github.com/haosulab/ManiSkill) | 🌟 [SAM3D](https://github.com/facebookresearch/sam-3d-objects) | 🌟 [infinigen](https://github.com/princeton-vl/infinigen)
 
 ---
 

apps/image_to_3d.py

@@ -83,7 +83,7 @@ def preprocess_example_image(
             <a href="https://github.com/HorizonRobotics/EmbodiedGen">
                 <img alt="💻 GitHub" src="https://img.shields.io/badge/GitHub-000000?logo=github">
             </a>
-            <a href="https://www.youtube.com/watch?v=rG4odybuJRk">
+            <a href="https://youtu.be/swjDFEvjn14">
                 <img alt="🎥 Video" src="https://img.shields.io/badge/🎥-Video-red">
             </a>
         </p>

apps/text_to_3d.py

@@ -63,7 +63,7 @@
             <a href="https://github.com/HorizonRobotics/EmbodiedGen">
                 <img alt="💻 GitHub" src="https://img.shields.io/badge/GitHub-000000?logo=github">
             </a>
-            <a href="https://www.youtube.com/watch?v=rG4odybuJRk">
+            <a href="https://youtu.be/swjDFEvjn14">
                 <img alt="🎥 Video" src="https://img.shields.io/badge/🎥-Video-red">
             </a>
         </p>

apps/texture_edit.py

@@ -64,15 +64,13 @@ def active_btn_by_content(mesh_content: gr.Model3D, text_content: gr.Textbox):
             <a href="https://github.com/HorizonRobotics/EmbodiedGen">
                 <img alt="💻 GitHub" src="https://img.shields.io/badge/GitHub-000000?logo=github">
             </a>
-            <a href="https://www.youtube.com/watch?v=rG4odybuJRk">
+            <a href="https://youtu.be/swjDFEvjn14">
                 <img alt="🎥 Video" src="https://img.shields.io/badge/🎥-Video-red">
             </a>
         </p>
 
         🎨 Generate visually rich textures for 3D mesh.
-        """.format(
-            VERSION=VERSION
-        ),
+        """.format(VERSION=VERSION),
         elem_classes=["header"],
     )
 

apps/visualize_asset.py

@@ -56,6 +56,13 @@
 HF_REPO_ID = "HorizonRobotics/EmbodiedGenData"
 HF_LOCAL_DIR = "EmbodiedGenData"
 CAMERA_ZOOM = 3.2
+# Gradio's default per-event concurrency limit is 1, which would make every
+# user's HF snapshot_download (potentially several seconds) serialize behind
+# whichever user clicked first. Give the HF-fetching events their own shared
+# pool so several users can download/preview assets in parallel, bounded to
+# avoid hammering the HF API with too many simultaneous requests.
+HF_IO_CONCURRENCY_ID = "hf_asset_io"
+HF_IO_CONCURRENCY_LIMIT = 6
 
 # Compatible with huggingface space zero GPU
 import spaces
@@ -124,6 +131,7 @@ def ensure_hf_files(rel_patterns: list[str]) -> None:
 css = """
 .gradio-container .gradio-group { box-shadow: 0 2px 4px rgba(0,0,0,0.05) !important; }
 #asset-gallery { border: 1px solid #E5E7EB; border-radius: 8px; padding: 8px; background-color: #F9FAFB; }
+#download-row button { white-space: nowrap; }
 """
 
 lighting_css = """
@@ -724,13 +732,33 @@ def download_mjcf(asset_dir: str, req: gr.Request) -> str | None:
     return zip_path
 
 
+def download_affordance(asset_dir: str, req: gr.Request) -> str | None:
+    """Package the ``affordance/`` folder."""
+    if not asset_dir or not os.path.isdir(asset_dir):
+        gr.Warning("Please select an asset first.")
+        return None
+
+    gr.Info("⏳ Preparing affordance asset for download, please wait...")
+    ensure_hf_files([f"{_rel_dir(asset_dir)}/affordance/**"])
+    affordance_dir = os.path.join(asset_dir, "affordance")
+    if not os.path.isdir(affordance_dir):
+        gr.Warning("No affordance folder available for this asset.")
+        return None
+
+    zip_path, uid, zip_name = _fmt_zip_path(asset_dir, "affordance", req)
+    _zip_items(zip_path, [(affordance_dir, "affordance")])
+    gr.Info(f"✅ {zip_name} is ready.")
+    return zip_path
+
+
 # Download buttons keyed by format; order matches the outputs list used when
 # locking/unlocking them together.
-_DL_KEYS = ("urdf", "usd", "mjcf")
+_DL_KEYS = ("urdf", "usd", "mjcf", "affordance")
 _DL_LABELS = {
     "urdf": "⬇️ Download URDF",
     "usd": "⬇️ Download USD",
     "mjcf": "⬇️ Download MJCF",
+    "affordance": "⬇️ Affordance",
 }
 
 
@@ -796,7 +824,7 @@ def end_session(req: gr.Request) -> None:
             <a href="https://github.com/HorizonRobotics/EmbodiedGen">
                 <img alt="💻 GitHub" src="https://img.shields.io/badge/GitHub-000000?logo=github">
             </a>
-            <a href="https://www.youtube.com/watch?v=rG4odybuJRk">
+            <a href="https://youtu.be/swjDFEvjn14">
                 <img alt="🎥 Video" src="https://img.shields.io/badge/🎥-Video-red">
             </a>
         </p>
@@ -931,7 +959,7 @@ def end_session(req: gr.Request) -> None:
                     urdf_file = gr.Textbox(
                         label="URDF File Path", interactive=False, lines=2
                     )
-                with gr.Row():
+                with gr.Row(elem_id="download-row"):
                     # DownloadButtons that build their zip into their own value
                     # on click; a chained JS handler then triggers the browser
                     # download from that value (one-click). If the JS ever
@@ -952,6 +980,11 @@ def end_session(req: gr.Request) -> None:
                         variant="primary",
                         interactive=False,
                     )
+                    affordance_dl_btn = gr.DownloadButton(
+                        label="⬇️ Affordance",
+                        variant="primary",
+                        interactive=False,
+                    )
 
     search_button.click(
         fn=search_assets,
@@ -1008,6 +1041,8 @@ def update_assets(p, s, c):
     gallery.select(
         fn=show_asset_from_gallery,
         inputs=[primary, secondary, category, search_box, gallery_df_state],
+        concurrency_id=HF_IO_CONCURRENCY_ID,
+        concurrency_limit=HF_IO_CONCURRENCY_LIMIT,
         outputs=[
             (visual_path_state := gr.State()),
             (collision_path_state := gr.State()),
@@ -1030,7 +1065,7 @@ def update_assets(p, s, c):
         ],
     ).success(
         fn=_reset_downloads,
-        outputs=[urdf_dl_btn, usd_dl_btn, mjcf_dl_btn],
+        outputs=[urdf_dl_btn, usd_dl_btn, mjcf_dl_btn, affordance_dl_btn],
     )
 
     # After the zip is built into the button's own value, pass that file value
@@ -1050,33 +1085,61 @@ def update_assets(p, s, c):
     }
     """
 
-    dl_btns = [urdf_dl_btn, usd_dl_btn, mjcf_dl_btn]
+    dl_btns = [urdf_dl_btn, usd_dl_btn, mjcf_dl_btn, affordance_dl_btn]
 
     # Each flow: lock all buttons (prevent spam clicks) -> build the zip ->
     # JS-trigger the browser download -> unlock. Selecting another asset
     # resets the buttons independently (see `.success` above).
     urdf_dl_btn.click(
-        fn=lambda: _lock_downloads("urdf"), outputs=dl_btns
+        fn=lambda: _lock_downloads("urdf"), outputs=dl_btns, queue=False
     ).then(
-        fn=download_urdf, inputs=[asset_folder], outputs=[urdf_dl_btn]
+        fn=download_urdf,
+        inputs=[asset_folder],
+        outputs=[urdf_dl_btn],
+        concurrency_id=HF_IO_CONCURRENCY_ID,
+        concurrency_limit=HF_IO_CONCURRENCY_LIMIT,
     ).then(fn=lambda *a: None, inputs=[urdf_dl_btn], js=download_js).then(
         fn=_unlock_downloads, outputs=dl_btns
     )
 
-    usd_dl_btn.click(fn=lambda: _lock_downloads("usd"), outputs=dl_btns).then(
-        fn=download_usd, inputs=[asset_folder], outputs=[usd_dl_btn]
+    usd_dl_btn.click(
+        fn=lambda: _lock_downloads("usd"), outputs=dl_btns, queue=False
+    ).then(
+        fn=download_usd,
+        inputs=[asset_folder],
+        outputs=[usd_dl_btn],
+        concurrency_id=HF_IO_CONCURRENCY_ID,
+        concurrency_limit=HF_IO_CONCURRENCY_LIMIT,
     ).then(fn=lambda *a: None, inputs=[usd_dl_btn], js=download_js).then(
         fn=_unlock_downloads, outputs=dl_btns
     )
 
     mjcf_dl_btn.click(
-        fn=lambda: _lock_downloads("mjcf"), outputs=dl_btns
+        fn=lambda: _lock_downloads("mjcf"), outputs=dl_btns, queue=False
     ).then(
-        fn=download_mjcf, inputs=[asset_folder], outputs=[mjcf_dl_btn]
+        fn=download_mjcf,
+        inputs=[asset_folder],
+        outputs=[mjcf_dl_btn],
+        concurrency_id=HF_IO_CONCURRENCY_ID,
+        concurrency_limit=HF_IO_CONCURRENCY_LIMIT,
     ).then(fn=lambda *a: None, inputs=[mjcf_dl_btn], js=download_js).then(
         fn=_unlock_downloads, outputs=dl_btns
     )
 
+    affordance_dl_btn.click(
+        fn=lambda: _lock_downloads("affordance"),
+        outputs=dl_btns,
+        queue=False,
+    ).then(
+        fn=download_affordance,
+        inputs=[asset_folder],
+        outputs=[affordance_dl_btn],
+        concurrency_id=HF_IO_CONCURRENCY_ID,
+        concurrency_limit=HF_IO_CONCURRENCY_LIMIT,
+    ).then(
+        fn=lambda *a: None, inputs=[affordance_dl_btn], js=download_js
+    ).then(fn=_unlock_downloads, outputs=dl_btns)
+
     demo.load(start_session)
     demo.unload(end_session)
 

docs/documentation/acknowledgement.md

@@ -1,7 +1,7 @@
 # 🙌 Acknowledgement
 
 EmbodiedGen builds upon the following amazing projects and models:
-🌟 [Trellis](https://github.com/microsoft/TRELLIS) | 🌟 [Hunyuan-Delight](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) | 🌟 [Segment Anything](https://github.com/facebookresearch/segment-anything) | 🌟 [Rembg](https://github.com/danielgatis/rembg) | 🌟 [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4) | 🌟 [Stable Diffusion x4](https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler) | 🌟 [Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) | 🌟 [Kolors](https://github.com/Kwai-Kolors/Kolors) | 🌟 [ChatGLM3](https://github.com/THUDM/ChatGLM3) | 🌟 [Aesthetic Score](http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html) | 🌟 [Pano2Room](https://github.com/TrickyGo/Pano2Room) | 🌟 [Diffusion360](https://github.com/ArcherFMY/SD-T2I-360PanoImage) | 🌟 [Kaolin](https://github.com/NVIDIAGameWorks/kaolin) | 🌟 [diffusers](https://github.com/huggingface/diffusers) | 🌟 [gsplat](https://github.com/nerfstudio-project/gsplat) | 🌟 [QWEN-2.5VL](https://github.com/QwenLM/Qwen2.5-VL) | 🌟 [GPT4o](https://platform.openai.com/docs/models/gpt-4o) | 🌟 [SD3.5](https://huggingface.co/stabilityai/stable-diffusion-3.5-medium) | 🌟 [ManiSkill](https://github.com/haosulab/ManiSkill) | 🌟 [SAM3D](https://github.com/facebookresearch/sam-3d-objects)
+🌟 [Trellis](https://github.com/microsoft/TRELLIS) | 🌟 [Hunyuan-Delight](https://huggingface.co/tencent/Hunyuan3D-2/tree/main/hunyuan3d-delight-v2-0) | 🌟 [Hunyuan3D-Part](https://github.com/Tencent-Hunyuan/Hunyuan3D-Part) | 🌟 [GraspGen](https://github.com/NVlabs/GraspGen) | 🌟 [Segment Anything](https://github.com/facebookresearch/segment-anything) | 🌟 [Rembg](https://github.com/danielgatis/rembg) | 🌟 [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4) | 🌟 [Stable Diffusion x4](https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler) | 🌟 [Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) | 🌟 [Kolors](https://github.com/Kwai-Kolors/Kolors) | 🌟 [ChatGLM3](https://github.com/THUDM/ChatGLM3) | 🌟 [Aesthetic Score](http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html) | 🌟 [Pano2Room](https://github.com/TrickyGo/Pano2Room) | 🌟 [Diffusion360](https://github.com/ArcherFMY/SD-T2I-360PanoImage) | 🌟 [Kaolin](https://github.com/NVIDIAGameWorks/kaolin) | 🌟 [diffusers](https://github.com/huggingface/diffusers) | 🌟 [gsplat](https://github.com/nerfstudio-project/gsplat) | 🌟 [QWEN-2.5VL](https://github.com/QwenLM/Qwen2.5-VL) | 🌟 [SD3.5](https://huggingface.co/stabilityai/stable-diffusion-3.5-medium) | 🌟 [ManiSkill](https://github.com/haosulab/ManiSkill) | 🌟 [SAM3D](https://github.com/facebookresearch/sam-3d-objects) | 🌟 [infinigen](https://github.com/princeton-vl/infinigen)
 
 ---
 

docs/documentation/index.md

@@ -9,7 +9,7 @@ hide:
 [![GitHub](https://img.shields.io/badge/GitHub-EmbodiedGen-black?logo=github)](https://github.com/HorizonRobotics/EmbodiedGen)
 [![📄 arXiv](https://img.shields.io/badge/📄-arXiv-b31b1b)](https://arxiv.org/abs/2506.10600)
 [![🤗 Dataset](https://img.shields.io/badge/🤗-Dataset-blue)](https://huggingface.co/datasets/HorizonRobotics/EmbodiedGenData)
-[![🎥 Video](https://img.shields.io/badge/🎥-Video-red)](https://www.youtube.com/watch?v=rG4odybuJRk)
+[![🎥 Video](https://img.shields.io/badge/🎥-Video-red)](https://youtu.be/swjDFEvjn14)
 
 *EmbodiedGen V2*: An Agentic, Simulation-Ready 3D World Engine for Embodied AI.
 

docs/documentation/install.md

@@ -38,5 +38,5 @@ Update the API key in file: `embodied_gen/utils/gpt_config.yaml`.
 
 You can choose between two backends for the GPT agent:
 
-- **`gpt-4o`** (Recommended) – Use this if you have access to **Azure OpenAI**.
-- **`qwen2.5-vl`** – An alternative with free usage via OpenRouter, apply a free key [here](https://openrouter.ai/settings/keys) and update `api_key` in `embodied_gen/utils/gpt_config.yaml` (50 free requests per day)
+- **`gpt-5.4`** (or Higher, Recommended) – Use this if you have access to **Azure OpenAI**.
+- **`gemma-4-31b`** – A free multimodal alternative (`google/gemma-4-31b-it:free`) via OpenRouter, apply a free key [here](https://openrouter.ai/settings/keys) and update `api_key` in `embodied_gen/utils/gpt_config.yaml`. Free-tier availability on OpenRouter can change over time — check the [model page](https://openrouter.ai/google/gemma-4-31b-it:free) before relying on it.