Foundation Models
Open-weight LLMs, fine-tuning stacks, and tooling around foundation models.
113 projects
See methodology for ranking rules; order uses public GitHub metrics within this scenario.
| Rank | Project | Stars | Forks | Updated | License |
|---|---|---|---|---|---|
| 1 |
transformers
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training. |
164.6K | 34.4K | 08/29/26 | Apache-2.0 |
| 2 |
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image |
34.2K | 4.0K | 03/25/26 | MIT |
| 3 |
MiniCPM-V
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone |
26.3K | 2.1K | 08/26/26 | Apache-2.0 |
| 4 |
generative-models
Generative Models by Stability AI |
27.3K | 3.1K | 12/16/25 | MIT |
| 5 |
unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities |
22.2K | 2.7K | 08/26/26 | MIT |
| 6 |
DeepSeek-Coder
DeepSeek Coder: Let the Code Write Itself |
24.2K | 2.9K | 11/11/25 | MIT |
| 7 |
Qwen
The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud. |
21.7K | 1.9K | 03/05/26 | Apache-2.0 |
| 8 |
Chinese-LLaMA-Alpaca
中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs) |
18.9K | 1.8K | 04/19/26 | Apache-2.0 |
| 9 |
RWKV-LM
RWKV is a parallelizable RNN with transformer-level LLM performance, offering linear time, constant space, and infinite context length. |
14.7K | 1.0K | 08/26/26 | Apache-2.0 |
| 10 |
open_clip
An open source implementation of CLIP. |
14.1K | 1.3K | 08/28/26 | Other |
| 11 |
LLaVA
[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond. |
25.0K | 2.8K | 08/12/24 | Apache-2.0 |
| 12 |
MiniCPM
MiniCPM5-1B: A SOTA 1B on-device LLM, small yet powerful. |
10.3K | 694 | 07/27/26 | Apache-2.0 |
| 13 |
needle
14MB foundation model for tiny devices; phones, wearables, smart home, and robots. |
9.7K | 622 | 08/29/26 | Apache-2.0 |
| 14 |
models
A collection of pre-trained, state-of-the-art models in the ONNX format |
9.8K | 1.6K | 08/01/26 | Apache-2.0 |
| 15 |
TabPFN
⚡ TabPFN: Foundation Model for Tabular Data ⚡ |
7.9K | 782 | 08/28/26 | Other |
| 16 |
Chinese-BERT-wwm
Pre-Training with Whole Word Masking for Chinese BERT（中文BERT-wwm系列模型） |
10.2K | 1.4K | 04/19/26 | Apache-2.0 |
| 17 |
Janus
Janus-Series: Unified Multimodal Understanding and Generation Models |
17.8K | 2.2K | 02/01/25 | MIT |
| 18 |
ERNIE
The official repository for ERNIE 4.5 and ERNIEKit – its industrial-grade development toolkit based on PaddlePaddle. |
7.7K | 1.4K | 07/24/26 | Apache-2.0 |
| 19 |
Kimi-K2
Kimi K2 is the large language model series developed by Moonshot AI team |
11.1K | 908 | 01/21/26 | Other |
| 20 |
GLM-5
GLM-5: From Vibe Coding to Agentic Engineering |
7.1K | 919 | 08/27/26 | Apache-2.0 |