#54 · Primary category: Foundation Models
xgen
Salesforce open-source LLMs with 8k sequence length.
Project last updated:06/02/26
GitHub Stars
726
Forks
40
Contributors
3
License
Apache-2.0
Why we included this project
XGen is the official research release of Salesforce's 7B parameter language model family, and it's a solid choice if you want a compact, Apache-licensed base model you can fine-tune or run yourself rather than calling a hosted API. The main draw is long-context handling: the base checkpoints support 4K and 8K input sequence lengths, which helps with summarizing long documents or reasoning over extended passages. The repo stays lean, pointing to Hugging Face model cards and including a short Transformers snippet that loads the tokenizer and model and generates text, so you can be up and running quickly. One caveat: the instruction-tuned variant is flagged for research purposes only, and the authors recommend evaluating accuracy, safety, and fairness before any production deployment.
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
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.
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
MiniCPM-V
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
generative-models
Generative Models by Stability AI
unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities