Search results

Consider using pre-built Flash Attention kernels via kernels

Status: Open.

#18 In jd-opensource/JoyAI-Image;

· sayakpaul opened on May 28, 2026

where is JoyAI-Image-Edit-Plus ??

Status: Open.

#16 In jd-opensource/JoyAI-Image;

· monnky opened on May 11, 2026

What‘s the right inference steps?

Status: Open.

#15 In jd-opensource/JoyAI-Image;

· rw6557079 opened on Apr 21, 2026

Directory structure mismatch in JoyAI-Image-Und (missing text_encoder/)

Status: Open.

#14 In jd-opensource/JoyAI-Image;

· fleph opened on Apr 21, 2026

官方的 ComfyUI 插件使用24G显存+32G共享内存是否能跑起来

Status: Open.

#13 In jd-opensource/JoyAI-Image;

· zhaojigang opened on Apr 17, 2026

Attention只能使用flash_attn方式计算吗？

Status: Open.

#12 In jd-opensource/JoyAI-Image;

· goldlee opened on Apr 15, 2026

如何配置以支持多参考图输入完成图像编辑的任务？

Status: Open.

#10 In jd-opensource/JoyAI-Image;

· whq1342 opened on Apr 11, 2026

The generated portrait images exhibit a distinct "AI look."

Status: Open.

#6 In jd-opensource/JoyAI-Image;

· hj13-mtlab opened on Apr 8, 2026

Batch inference support

Status: Open.

#5 In jd-opensource/JoyAI-Image;

· quetzalsly opened on Apr 8, 2026

clip编码模型用的是什么呢

Status: Open.

#4 In jd-opensource/JoyAI-Image;

· dxhark opened on Apr 8, 2026

很棒的工作，什么时候会开源训练lora的代码呢？

Status: Open.

#3 In jd-opensource/JoyAI-Image;

· willneverwill opened on Apr 6, 2026

Any distilled weights for fewer-step inference?

Status: Open.

#2 In jd-opensource/JoyAI-Image;

· vibe-viscot opened on Apr 4, 2026