“Latency–Throughput Tradeoffs of ONNX Runtime, TensorRT-LLM, VLLM, and Triton: An Empirical Comparison on 1B–3B Parameter LLM Inference”. 2026. Journal of Global Engineering Review 4 (1): 173-82. https://doi.org/10.66372/JGER.v4i1.12.