#71 · Primary category： MLOps & Evaluation
cube-studio
Open-source cloud-native AI platform for end-to-end MLOps, covering deep learning and large model training, inference, pipeline orchestration, and resource management.
Project last updated：07/11/26
GitHub Stars
5.1K
Forks
881
Contributors
40
License
Other
Why we included this project
Teams that run frequent training jobs on Kubernetes tend to accumulate a pile of separate tools, one for notebooks, one for pipelines, one for serving. Cube Studio tries to be the single platform that replaces all of them: it bundles online notebooks, a drag-and-drop pipeline editor, multi-node multi-GPU distributed training, hyperparameter search, and inference serving with VGPU virtualization. It has also grown to cover LLM work, including fine-tuning and reinforcement learning for DeepSeek-class models, multi-machine inference through vLLM and Ollama, private knowledge bases, a model marketplace, and support for Ascend NPUs and RDMA. That breadth makes it a plausible self-hosted alternative to commercial ML platforms if you want everything in one deployable package. One caveat for anyone considering it: the repository was archived in May 2026 and development has moved to the data-infra/cube-studio fork, so that is the repo to start from.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
unsloth
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, FLUX and more.
LlamaFactory
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
airflow
Apache Airflow - A platform to programmatically author, schedule, and monitor workflows
langfuse
🪢 Open source AI engineering platform: LLM evals, observability, metrics, prompt management, playground, datasets. Integrates with OpenTelemetry, LangChain, OpenAI SDK, LiteLLM, and more. 🍊YC W23
netron
Visualizer for neural network, deep learning and machine learning models