在哪里可以试用多轮对话生成图片？

[chenbaiyujason]

[wangyang0210]

https://hunyuan.tencent.com/bot/chat v1直接混元就行了

https://image.hunyuan.tencent.com/authorize

v2版本需申请排队

[wangyang0210]

[yestinl]

https://hunyuan.tencent.com/bot/chat v1直接混元就行了

https://image.hunyuan.tencent.com/authorize

v2版本需申请排队

Thanks @wangyang0210 for the reply! That's correct. You can directly use multi-turn text interactions to generate the image by Tencent Hunyuan Bot. We will also provide a multi-turn T2I (Text-to-Image) generation Gradio demo soon.

[Cerf-Volant425]

https://hunyuan.tencent.com/bot/chat v1直接混元就行了 https://image.hunyuan.tencent.com/authorize v2版本需申请排队

Thanks @wangyang0210 for the reply! That's correct. You can directly use multi-turn text interactions to generate the image by Tencent Hunyuan Bot. We will also provide a multi-turn T2I (Text-to-Image) generation Gradio demo soon.

Thanks for your great work. I only found the instructions for text2img. And I noticed you hava already uploaded the dialoggen model to the huggingface, I was wondering whether there are instructions for multi-round dialogue of this model. Or I should wait for the test of hunyuan_v2.

[yestinl]

@Cerf-Volant425 As described in Readme, you can modify the image through continuous language interaction in Tencent Hunyuan Bot. Here's an example: .

Thank you for your suggestions. We will update a demo that can run multi-turn T2I generation later.

[yestinl]

@wangyang0210 @chenbaiyujason 最新更新，我们提供了多轮对话绘图的demo.你可以通过以下命令来运行： (Latest news, we provide a demo of multi-turn dialogue drawing. You can run it with the following command:)

# Start a multi-turn T2I generation UI. 
# If your GPU memory is less than 32GB, use '--load-4bit' to enable 4-bit quantization, which requires at least 22GB of memory.
python app/multiTurnT2I_app.py

如果有任何其他问题，随时可以重新开启这个issue. (If you have any other questions, feel free to reopen this issue at any time.)