SEED

SEED copied to clipboard

From the result of magvit3 image tokenizer, it's reconstruct result far better than SEED tokenizer.

lucasjinreal

Hi，I think this is a nice work,i want to know What is the maximum number of tokens your model can handle as input?

Caixy1113

Dears, Thanks for your great work, and thanks for sharing it! Unfortunately, the shared training codes are very hard to run. The repo contains many errors, e.g., import errors due...

eslambakr

Thanks for your great work! I am wondering if you can provide the evaluation code for performing image-text/text-image retrieval as done in Table 1.

SxJyJay

I've noticed you mentioned interleaved text-image generation only able on instruction-tuned model, would you release in the future? or would you please release some methods of instruction-tuning?

URRealHero