AUR (en) - python-ai-dynamo

Search Criteria

Enter search criteria

Search by

Keywords

Out of Date

Sort by

Sort order

Per page

Git Clone URL:	https://aur.archlinux.org/python-ai-dynamo.git (read-only, click to copy)
Package Base:	python-ai-dynamo
Description:	NVIDIA Dynamo: datacenter-scale distributed inference serving framework (Python)
Upstream URL:	https://github.com/ai-dynamo/dynamo
Licenses:	Apache-2.0
Submitter:	Smoolak
Maintainer:	Smoolak
Last Packager:	Smoolak
Votes:	1
Popularity:	0.58
First Submitted:	2026-06-01 11:02 (UTC)
Last Updated:	2026-06-01 11:02 (UTC)

python
python-ai-dynamo-runtime^AUR
python-aiohttp
python-kubernetes (python-kubernetes-git^AUR)
python-msgspec
python-prometheus_client
python-pyzmq
python-transformers^AUR (python-transformers-git^AUR)
git (git-git^AUR, git-gl^AUR, git-wd40^AUR) (make)
python-build (make)
python-hatchling (python-hatchling-git^AUR) (make)
python-installer (make)
python-wheel (make)
python-nixl^AUR (optional) – NIXL data-transfer connector (dynamo.nixl_connect, multimodal KV transfer)
python-pmdarima^AUR (optional) – load forecasting for the SLA planner (dynamo.planner)
python-sglang (optional) – SGLang inference backend (dynamo.sglang)
python-tensorrt-llm^AUR (optional) – TensorRT-LLM inference backend (dynamo.trtllm)
python-uvloop (optional) – faster asyncio event loop used by the inference backends
python-vllm^AUR (python-vllm-bin^AUR, python-vllm-cuda^AUR, python-vllm-rocm^AUR) (optional) – vLLM inference backend (dynamo.vllm)