| High-power AI launcher | |
| Recommended immediate ONNX model: | |
| - onnx-community/Hermes-4-14B-Onnx | |
| Bigger base models to consider (not bundled here): | |
| - Qwen/Qwen2.5-72B-Instruct | |
| - meta-llama/Llama-3.1-405B-Instruct | |
| Notes: | |
| - ONNX Runtime is for inference/deployment. | |
| - Training/fine-tuning should be done in a framework first. | |
| - If you want to export a smaller PyTorch model to ONNX, use Hugging Face Optimum or torch.onnx. |