tensorchord/awesome-llmops

Awesome LLMOps is a curated reference list for developers working with large language models in production. LLMOps, short for Large Language Model Operations, refers to the tooling and processes involved in training, deploying, monitoring, and maintaining AI language models as working software systems. This repository collects links to relevant open-source projects and organizes them by category so developers can find what they need without searching across many sources separately. The list is organized into broad sections. The Model section links to notable open-weight language models, including text models, image and video foundation models, and audio models. The Serving section covers tools for deploying models at scale, from inference servers to frameworks that handle batching and optimization. Security and observability tools for monitoring model behavior in production get their own section. There are also dedicated sections for vector search databases, Code AI tools, and training infrastructure including experiment tracking and fine-tuning frameworks. Further sections cover data management for machine learning pipelines, including storage backends, feature engineering tools, and data labeling systems. Large-scale deployment infrastructure such as ML platforms, workflow orchestration, and model registries are listed separately. A performance section covers compilers and profiling tools for squeezing speed out of model inference. AutoML, optimization techniques, and federated learning each have their own groupings. Each entry in the list includes a short description and a GitHub star count badge where available. Most entries link directly to the referenced open-source project. The list accepts community contributions, with a contributing guide linked in the README. A Discord server is available for discussion. The full README is longer than what was shown.