AILab-CVC/SEED
Multimodal LLM combining vision and language capabilities, implemented with training code for pretraining and instruction tuning.
Not currently ranked — collecting fresh signals.
star history
This repository provides the official implementation of SEED-LLaMA, a multimodal large language model that integrates visual and textual understanding. The codebase includes the SEED tokenizer, multimodal LLM pretraining pipeline, and instruction tuning components. It supports large-scale multi-node training with DeepSpeed and provides efficient training data pipelines for vision-language model development.
Frequently asked
- What is AILab-CVC/SEED?
- Multimodal LLM combining vision and language capabilities, implemented with training code for pretraining and instruction tuning.
- Is SEED open source?
- Yes — AILab-CVC/SEED is an open-source project tracked on heatdrop.
- What language is SEED written in?
- AILab-CVC/SEED is primarily written in Python.
- How popular is SEED?
- AILab-CVC/SEED has 642 stars on GitHub.
- Where can I find SEED?
- AILab-CVC/SEED is on GitHub at https://github.com/AILab-CVC/SEED.