#5555th of 105 in ML frameworks and libraries
modelscope
Python library for running and fine-tuning ModelScope Hub models
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 9.2k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
8,762 on 1 Apr 20269,180 on 21 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
The core ModelScope Python library gives a unified `pipeline` interface for inference and a `Trainer` interface for fine-tuning and evaluation across NLP, CV, audio, multi-modal and science models. It also handles model and dataset download, versioning and caching against the ModelScope Hub. It runs on PyTorch, TensorFlow or ONNX, with per-domain pip extras.
Who it is for: Python developers running or fine-tuning ModelScope Hub models
Strengths
- Inference in a few lines via `pipeline`; fine-tuning via `build_trainer` and `trainer.train()`
- Covers NLP, CV, audio, multi-modal and AI-for-science tasks under one API
- Built-in Model-Hub and Dataset-Hub access with version control and cache management
- Official CPU and GPU Docker images published for each release
Weaknesses
- README examples and Docker tags reference old stacks (Python 3.7/3.8, TF 1.15, torch 1.11/2.0.1)
- Some audio models only support Python 3.7 and TensorFlow 1.15.4 on Linux
- Vision models may need mmcv-full; audio models need libsndfile on Linux
- Model catalog and many docs links center on modelscope.cn, in Chinese
What it needs
- GPU optional
- Docker
- Needs PyTorch, TensorFlow, ONNX, mmcv-full, libsndfile, ModelScope Hub
- Models: PyTorch, TensorFlow, ONNX, LLM, multi-modal
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |