diff --git a/README.md b/README.md
index e46c75a..3ec9605 100644
--- a/README.md
+++ b/README.md
@@ -37,6 +37,8 @@ The toolbox supports the most advanced high-quality navigation dataset, InternDa
 ## 🔥 News
 | Time   | Update |
 |---------|--------|
+| 2025/12 | InternNav v0.3.0 released. |
+| 2025/12 | We introduce Interactive Instance Goal Navigation (IIGN) and release VL-LN Bench to enable InternVLA-N1 to solve this task, with large-scale dialog-trajectory collection plus training and evaluation support. See [our website](https://0309hws.github.io/VL-LN.github.io/) for details.|
 | 2025/12 | Training code for InternVLA-N1 is now available. This release provides two model configurations: InternVLA-N1 (Dual System)<span style="color: #28a745; font-size: 0.9em"> with NavDP*</span> and InternVLA-N1 (Dual System)<span style="color: #28a745; font-size: 0.9em"> DualVLN </span>. For model architecture and training details, please refer to the [DualVLN paper](https://arxiv.org/abs/2512.08186).|
 | 2025/11 | InternNav v0.2.0 released — added distributed evaluation support for VLN-PE.|
 | 2025/10 | Add a [inference-only demo](scripts/notebooks/inference_only_demo.ipynb) of InternVLA-N1. |
diff --git a/setup.py b/setup.py
index 9d3631c..8b94f51 100644
--- a/setup.py
+++ b/setup.py
@@ -55,7 +55,7 @@ def parse_readme(readme: str) -> str:
 
 setuptools.setup(
     name='internnav',
-    version='0.2.0',
+    version='0.3.0',
     packages=setuptools.find_packages(),
     author='Intern Robotics',
     author_email='embodiedai@pjlab.org.cn',
@@ -78,11 +78,6 @@ def parse_readme(readme: str) -> str:
         # envs
         "isaac": isaac_requires,
         "habitat": habitat_requires,
-        "demo": [
-            "gradio==5.45",
-            "hf-xet==1.1.5",
-            "huggingface-hub==0.33.4",
-        ],
         # models
         "internvla_n1": n1_requires,
         "baseline": model_requires,