| --- |
| license: other |
| license_name: embedl-software-2.1 |
| license_link: https://embedl.com/legal/software-licence/v2.1 |
| base_model: nvidia/Alpamayo2-Edge |
| pipeline_tag: robotics |
| tags: |
| - autonomous-driving |
| - vision-language-action |
| - alpamayo |
| - jetson-thor |
| - tensorrt-edgellm |
| - nvfp4 |
| - mxfp8 |
| - embedl |
| --- |
| |
| # Alpamayo 2 Edge — 92 ms on Jetson AGX Thor |
|
|
| **10.9 decisions/s on NVIDIA Jetson AGX Thor: 6.6× faster than NVIDIA's PyTorch BF16, with lower trajectory error.** |
|
|
| <!-- embedl-run-stickers:start --> |
| <p class="dark:hidden not-prose" style="margin:8px 0 16px 0;"> |
| <a href="https://embedl.com/nvidia/alpamayo2-edge/thor?tab=run&utm_source=huggingface&utm_medium=model-card&utm_campaign=run-sticker&utm_content=embedl%2FAlpamayo2-Edge" title="Run Alpamayo 2 Edge on AGX Thor: 92 ms, free on Embedl's device farm, embedl.com"><img src="https://storage.googleapis.com/embedl-model-media/sha256/734cba61c34690a878e4c9e908e767bafd65747b36f371e9e97c98d5d29e00d8.png" alt="Run on AGX Thor: 92 ms. Free on Embedl's device farm (embedl.com)" width="270" height="108" style="display:inline-block;margin:0 2px 4px 0;"></a> |
| </p> |
| <p class="hidden dark:block not-prose" style="margin:8px 0 16px 0;"> |
| <a href="https://embedl.com/nvidia/alpamayo2-edge/thor?tab=run&utm_source=huggingface&utm_medium=model-card&utm_campaign=run-sticker&utm_content=embedl%2FAlpamayo2-Edge" title="Run Alpamayo 2 Edge on AGX Thor: 92 ms, free on Embedl's device farm, embedl.com"><img src="https://storage.googleapis.com/embedl-model-media/sha256/3dd3d0a4af0b47600282a0586cb45a96d610be6c9e3ddf6f596795fc227263d4.png" alt="Run on AGX Thor: 92 ms. Free on Embedl's device farm (embedl.com)" width="270" height="124" style="display:inline-block;margin:0 2px 4px 0;"></a> |
| </p> |
| <!-- embedl-run-stickers:end --> |
|
|
| > **▶ Run Alpamayo 2 Edge live on a real Jetson AGX Thor — free, in your browser: |
| > [embedl.com/nvidia/alpamayo2-edge/thor](https://embedl.com/nvidia/alpamayo2-edge/thor?tab=run&utm_source=huggingface&utm_medium=model-card&utm_campaign=run-cta&utm_content=embedl%2FAlpamayo2-Edge)** |
|
|
| NVIDIA's [Alpamayo 2 Edge](https://huggingface.co/nvidia/Alpamayo2-Edge), the |
| vision-language-action model for autonomous driving, quantized and optimized |
| by Embedl to run in real time on NVIDIA Jetson AGX Thor with TensorRT Edge-LLM, |
| using 4.2× less energy per decision than NVIDIA's PyTorch BF16 release on the |
| same device. |
|
|
| <video src="https://storage.googleapis.com/embedl-model-media/sha256/6f0f8172b7b658e95efef45add6f18c5b46ec3dd78d518dcbd627f0b74c3ee34.mp4" controls muted playsinline width="100%"></video> |
|
|
| ## Run it live |
|
|
| Try the optimized model on a real Jetson AGX Thor in Embedl's device farm, for |
| free, with no hardware or setup: |
| **[Run Alpamayo 2 Edge on AGX Thor →](https://embedl.com/nvidia/alpamayo2-edge/thor?tab=run&utm_source=huggingface&utm_medium=model-card&utm_campaign=run-cta&utm_content=embedl%2FAlpamayo2-Edge)** |
|
|
| ## Get the model |
|
|
| The model and its runtime are distributed as container images from Embedl: |
| see **[Alpamayo 2 Edge on embedl.com](https://embedl.com/nvidia/alpamayo2-edge)** |
| for the downloads and instructions. |
|
|
| ## Quantization |
|
|
| | Component | Precision | |
| |---|---| |
| | Language model | NVFP4 (driving-calibrated), FP16 `lm_head` | |
| | Vision encoder (ViT) | MXFP8 | |
| | Action expert | MXFP8 | |
|
|
| Built from NVIDIA's released weights, `nvidia/Alpamayo2-Edge` at `b92b6aa3`. |
| Runtime: TensorRT Edge-LLM 0.10.1 with Embedl optimizations. |
|
|
| ## Results on Jetson AGX Thor |
|
|
| | Metric | Embedl NVFP4 / MXFP8 | TensorRT Edge-LLM FP16 | NVIDIA PyTorch BF16 | |
| |---|---:|---:|---:| |
| | Latency per decision, median | **91.9 ms** | 489.3 ms | 603.7 ms | |
| | Latency per decision, p90 | **140.3 ms** | | 840.6 ms | |
| | Decisions per second | **10.9 Hz** | | 1.7 Hz | |
| | Energy per decision | **12.45 J** | 62.02 J | 51.87 J | |
| | Power while deciding, mean | 117.6 W | | 74.5 W | |
| | Power while deciding, peak | 150.2 W | | 137.9 W | |
| | Model size | **2.37 GiB** | | 4.95 GiB | |
| | Runtime container size | **1.07 GiB** | | 11.35 GiB | |
| | minADE₆ at 5 s (lower is better) | **0.464 m** | | 0.508 m | |
|
|
| Board power at idle: 30.0 W. |
|
|
| ### Methodology |
|
|
| - **Latency:** one PhysicalAI-AV clip, all 180 decisions (one per 0.1 s of |
| driving), run one after another on the device. A decision is the model's |
| whole call from camera frames to trajectory, with the inputs on the GPU. |
| Decisions per second is 1000 / the median. |
| - **Accuracy:** minADE₆ at 5 s, the closest of 6 sampled trajectories, |
| average displacement over the first 5 s, on 378 held-out PhysicalAI-AV |
| validation clips, 2 seeds. The Embedl build is measured on the device; |
| NVIDIA's BF16 on an H200 with NVIDIA's own code, on the same events and |
| seeds. |
| - **Power:** the board's input power (tegrastats, every 500 ms) sampled |
| during the 180 timed decisions; energy per decision is the mean power times |
| the mean decision time. |
| - **Size:** the runtime container image and the model files on disk. |
| - **TensorRT Edge-LLM FP16:** the same release in FP16 on the same runtime and |
| device, without Embedl's quantization, caches or speculation. |
| - **NVIDIA PyTorch BF16:** NVIDIA's released code (NVlabs/alpamayo2-edge) and |
| BF16 weights with its `--accel` tuned path, on the same device and inputs. |
|
|
| ## License |
|
|
| This optimized model is licensed under the |
| [Embedl Software Licence 2.1](https://embedl.com/legal/software-licence/v2.1). |
| It is derived from [nvidia/Alpamayo2-Edge](https://huggingface.co/nvidia/Alpamayo2-Edge); |
| NVIDIA's original weights remain under their own licence. |
|
|
