Triton-hf-builder: Generate NVIDIA Triton repositories from Hugging Face checkpoints

An open-source Python library and CLI called triton-hf-builder has been released to generate NVIDIA Triton model repositories from local Hugging Face checkpoints, producing model configuration, runtime adapters and dependency files for Transformers, vLLM and TensorRT-LLM. Repository generation runs without a GPU and does not load model weights, and the project ships a Docker quickstart that downloads a small checkpoint, starts CPU serving and sends an inference request. The tool is in beta, licensed under Apache-2.0, and installs via pip; model and hardware compatibility depends on the selected serving backend.

Hi everyone I’m sharing triton-hf-builder , an open-source Python library and CLI that generates NVIDIA Triton model repositories from local Hugging Face checkpoints. It creates model configuration, runtime adapters and dependency files for Transformers, vLLM and TensorRT-LLM . Repository generation runs without a GPU and does not load model weights. The project includes a Docker quickstart that downloads a small checkpoint, starts CPU serving and sends an inference request. pip install triton-hf-builder Currently in beta, licensed under Apache-2.0. Model and hardware compatibility depends on the selected serving backend. Feedback, bug reports and experiences with different model architectures are welcome