OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models
This provides an open-source alternative for vision-language modeling, but it is incremental as it replicates existing work.
The authors tackled the problem of replicating DeepMind's Flamingo vision-language models by introducing OpenFlamingo, an open-source framework with models from 3B to 9B parameters, achieving 80-89% of Flamingo's performance on seven datasets.
We introduce OpenFlamingo, a family of autoregressive vision-language models ranging from 3B to 9B parameters. OpenFlamingo is an ongoing effort to produce an open-source replication of DeepMind's Flamingo models. On seven vision-language datasets, OpenFlamingo models average between 80 - 89% of corresponding Flamingo performance. This technical report describes our models, training data, hyperparameters, and evaluation suite. We share our models and code at https://github.com/mlfoundations/open_flamingo.