We propose Index Backpropagation Quantization (IBQ), a new vector quantization method for the joint optimization of all codebook embeddings and the visual encoder, ensuring the consistent latent space. IBQ enables scalable training of visual tokenizers and, for the first time, achieves a large-scale codebook (2^18) with high dimension (256) and high utilization.
Runs of TencentARC IBQ-Tokenizer-1024 on huggingface.co
6
Total runs
0
24-hour runs
-1
3-day runs
0
7-day runs
-2
30-day runs
More Information About IBQ-Tokenizer-1024 huggingface.co Model
IBQ-Tokenizer-1024 huggingface.co is an AI model on huggingface.co that provides IBQ-Tokenizer-1024's model effect (), which can be used instantly with this TencentARC IBQ-Tokenizer-1024 model. huggingface.co supports a free trial of the IBQ-Tokenizer-1024 model, and also provides paid use of the IBQ-Tokenizer-1024. Support call IBQ-Tokenizer-1024 model through api, including Node.js, Python, http.
IBQ-Tokenizer-1024 huggingface.co is an online trial and call api platform, which integrates IBQ-Tokenizer-1024's modeling effects, including api services, and provides a free online trial of IBQ-Tokenizer-1024, you can try IBQ-Tokenizer-1024 online for free by clicking the link below.
TencentARC IBQ-Tokenizer-1024 online free url in huggingface.co:
IBQ-Tokenizer-1024 is an open source model from GitHub that offers a free installation service, and any user can find IBQ-Tokenizer-1024 on GitHub to install. At the same time, huggingface.co provides the effect of IBQ-Tokenizer-1024 install, users can directly use IBQ-Tokenizer-1024 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.