ANOLE: An Open, Autoregressive, Native Large Multimodal Models for Interleaved Image-Text Generation
ethanchern
Abstract
Previous open-source large multimodal models (LMMs) have faced several limitations: (1) they often lack native integration, requiring adapters to align visual representations with pre-trained large language models (LLMs); (2) many are restricted to single-modal generation; (3) while some support multimodal generation, they rely on separate diffusion models for visual modeling and generation. To mitigate these limitations, we present Anole, an open, autoregressive, native large multimodal model for interleaved image-text generation. We build Anole from Meta AI's Chameleon, adopting an innovative fine-tuning strategy that is both data-efficient and parameter-efficient. Anole demonstrates high-quality, coherent multimodal generation capabilities. We have open-sourced our model, training framework, and instruction tuning data.
Community
Abstract
Previous open-source large multimodal models (LMMs) have faced several limitations: (1) they often lack native integration, requiring adapters to align visual representations with pre-trained large language models (LLMs); (2) many are restricted to single-modal generation; (3) while some support multimodal generation, they rely on separate diffusion models for visual modeling and generation. To mitigate these limitations, we present Anole, an open, autoregressive, native large multimodal model for interleaved image-text generation. We build Anole from Meta AI's Chameleon, adopting an innovative fine-tuning strategy that is both data-efficient and parameter-efficient. Anole demonstrates high-quality, coherent multimodal generation capabilities. We have open-sourced our model, training framework, and instruction tuning data.
Homepage: https://gair-nlp.github.io/anole
Code: https://github.com/GAIR-NLP/anole
Hugging Face Model: https://huggingface.co/GAIR/Anole-7b-v0.1
Hi @ethanchern congrats on this work! Some remarks:
- Would you be able to link the HF model to this paper page? See here on how to do that: https://huggingface.co/docs/hub/en/model-cards#linking-a-paper
- Currently downloads aren't working for your model, this is because the model is part of a "model" folder rather than being at the root of the repo. See here regarding how you can make downloads work: https://huggingface.co/docs/hub/models-download-stats
Let me know if you need any help!
Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper