Examples

End-to-end workflows demonstrating TensorRT Edge-LLM capabilities across different use cases.

Available Examples

• VLM (Vision-Language Model) - Complete workflow for vision-language models with image understanding capabilities

• Speculative Decoding - EAGLE speculative decoding for faster inference

• Phi-4 Multimodal - Phi-4-Multimodal deployment with LoRA merge

• ASR (Automatic Speech Recognition) - Speech-to-text with Qwen3-ASR models

• MoE (Mixture of Experts) - Mixture of Experts model deployment

• TTS (Text-to-Speech) - Text-to-speech synthesis workflows

• Experimental High-Level Python API and Server - vLLM-style API and OpenAI-compatible server with spec-decode support