Disty0
### Describe the bug Wuerstchen and Wuerstchen based Stable Cascade models generates corrupted images. - Happens with any resolution. - Happens with both Intel ARC and Intel Datacenter GPU Max....
### Describe the bug This bug happens with any Diffusion model. Meaning it's not specific to Stable Diffusion models. - Only happens at 1024x1024 or 1024x1024 sized (2048x512 etc) resolutions....
Is it possible to decrease the average block time from 15 seconds to around 1 second? If so dropping the block reward from 7 to 0.5 is also needed. 15...
torch.Generator is not supported on Intel so we have to run BrownianTreeNoiseSampler on the CPU with transform and send results back to the GPU with transform_last. It will throw this...
# What does this PR do? Fixes NNCF compression with Stable Cascade. NNCF compresses model weight to 8 bit and reduces the model footprint by half. Full Stable Cascade model...
# What does this PR do? Fixes # ([issue](https://github.com/huggingface/optimum-quanto/issues/182)) GEMM and GEMV kernels can't be compiled for AMD HIP. This PR adds a check for HIP and uses unpack kernel...
Fixes Python 3.13 compatibility. Closes https://github.com/XPixelGroup/BasicSR/issues/725 exec doesn't update locals with Python 3.13. This PR parses the version file as a text instead of running exec. BasicSR cannot be installed...
### Describe the bug IPEX 2.5 XPU and 2.6 XPU cannot be imported on latest Arch Linux. IPEX 2.3 and below and pure PyTorch 2.5 / 2.6 / 2.7 XPU...
Fixes Python 3.13 compatibility. exec doesn't update locals with Python 3.13. This PR parses the version file as a text instead of running exec. GFPGAN cannot be installed with Python...
# What does this PR do? Adds NewbieAI support to Diffusers. Adds `pooled_projection_dim` config to Lumina2Transformer2DModel and uses pooled projections from Newbie codebase if it is set to something other...