#80 · Primary category： Foundation Models
TigerBot
TigerBot: A multi-language multi-task LLM
Project last updated：12/28/24
GitHub Stars
2.3K
Forks
188
Contributors
19
License
Apache-2.0
Why we included this project
TigerBot is a family of bilingual Chinese-English models that continue pretraining from the Llama-2 checkpoints, with an expanded Chinese vocabulary and extra training tokens covering Chinese, code, and reasoning. Sizes run from a 7B base up to a 180B variant, so teams can pick a scale that fits their hardware rather than being locked into one model. The repository ships more than just weights: the training recipes are there, along with the alignment methods used (supervised fine-tuning, rejection sampling, DPO) and an OpenCompass-based evaluation setup that makes the results easier to reproduce. That makes it a practical starting point for teams building Chinese-language products on an open model, while the released technical report gives researchers a closer look at the data choices behind each version.
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