docs: improve readme

Author: shuheng-liu

README.md

@@ -19,17 +19,18 @@ Whether you use the official OpenPi codebase or LeRobot's reimplementation, you
 
 OpenTau ($\tau$) is a tool developed by *[Tensor](https://www.tensor.auto)* to bridge this gap. We want to help everyone train VLAs on their own datasets while fully leveraging the key techniques that make the $\pi$ models so powerful.
 
-|                                                 Features |          OpenPi           |             LeRobot              | **OpenTau** |
-|---------------------------------------------------------:|:-------------------------:|:--------------------------------:|:-----------:|
-|                  Co-training with Heterogeneous Datasets |             ❌             |                ❌                 |      ✅      |
-|                  Discrete Actions Training in $\pi_{0.5}$|             ❌             |                ❌                 |      ✅      |
-| Knowledge Insulation (KI) between VLM and Action Decoder |             ❌             |                ❌                 |      ✅      |
-|                              Dropout Layers in PaliGemma | ✅ (Jax) <br>❌ (PyTorch) |                ❌                 |      ✅      |
-|                        Multi-Node and Multi-GPU Training |             ❌             |                ✅                 |      ✅      |
-|                 Fully Functioning $\pi_{0.5}$ Checkpoint |             ✅             | ❌ <br> (Missing Text Embeddings) |      ✅      |
-|            Simulation Environments for Evaluating Models |             ❌             |                ✅                 |      ✅      |
-|    $\pi^{*}_{0.6}$ style Reinforcement Learning Pipeline |             ❌             |                ❌                 |      ✅      |
-|                                                Framework |       Jax / PyTorch       |             PyTorch              |   PyTorch   |
+|                                                         Features |          OpenPi           |             LeRobot              | **OpenTau** |
+|-----------------------------------------------------------------:|:-------------------------:|:--------------------------------:|:-----------:|
+|                          Co-training with Heterogeneous Datasets |             ❌             |                ❌                 |      ✅      |
+|                         Discrete Actions Training in $\pi_{0.5}$ |             ❌             |                ❌                 |      ✅      |
+|         Knowledge Insulation (KI) between VLM and Action Decoder |             ❌             |                ❌                 |      ✅      |
+|                                      Dropout Layers in PaliGemma | ✅ (Jax) <br>❌ (PyTorch) |                ❌                 |      ✅      |
+|                                Multi-Node and Multi-GPU Training |             ❌             |                ✅                 |      ✅      |
+|                         Fully Functioning $\pi_{0.5}$ Checkpoint |             ✅             | ❌ <br> (Missing Text Embeddings) |      ✅      |
+|                    Simulation Environments for Evaluating Models |             ❌             |                ✅                 |      ✅      |
+| Async Vector Sim with Grid Summary and Colored Success Indicator |             ❌             |                ❌                 |      ✅      |
+|            $\pi^{*}_{0.6}$ style Reinforcement Learning Pipeline |             ❌             |                ❌                 |      ✅      |
+|                                                        Framework |       Jax / PyTorch       |             PyTorch              |   PyTorch   |
 
 ## Quick Start
 If you are familiar with LeRobot, getting started with OpenTau is very easy.