modelscope/ms-swift
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.5, DeepSeek-R1, GLM-5, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Llava, Phi4, ...) (AAAI 2025).
This framework helps AI developers and researchers customize large language models (LLMs) and multimodal large models (MLLMs) for specific tasks or datasets. It takes a base model and your specialized data, then outputs a fine-tuned model ready for deployment. This is for professionals building custom AI applications.
13,105 stars. Used by 1 other package. Actively maintained with 89 commits in the last 30 days. Available on PyPI.
Use this if you need to fine-tune existing large language or multimodal models to perform better on your unique data or specialized tasks, across various hardware and with advanced optimization techniques.
Not ideal if you are looking for a pre-trained model to use directly, or if you do not have specific data to fine-tune a model.
Stars
13,105
Forks
1,255
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 13, 2026
Commits (30d)
89
Dependencies
40
Reverse dependents
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/modelscope/ms-swift"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Recent Releases
Related models
unslothai/unsloth
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek, Qwen, Llama,...
huggingface/peft
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
oumi-ai/oumi
Easily fine-tune, evaluate and deploy gpt-oss, Qwen3, DeepSeek-R1, or any open source LLM / VLM!
linkedin/Liger-Kernel
Efficient Triton Kernels for LLM Training
hiyouga/LlamaFactory
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)