Skip to content
代码片段 群组 项目
选择Git版本
20 个结果
可以使用方向键移动图形。
Created with Raphaël 2.2.027Nov262524232221201918171615141312111098765432131Oct30292827262524232221201918171615141312111098765432130Sep292827262524252423222122212019181716151413141312111098765432131Aug30292827262524232221201918171615141312111098765432131Julnitunion_typesunion_typesMove codeAdd union type supportLTXV model merging node.Merge branch 'master' into patch_hooks_improved_memoryAlternative fix for #5767fix multi add makedirs error (#5786)Support official SD3.5 Controlnets.v0.3.5v0.3.5Support for the official mochi lora format.Flux inpaint model does not work in fp16.Add option to inference the diffusion model in fp32 and fp64.Merge branch 'master' into patch_hooks_improved_memoryAdd changes manually from 'master' so merge conflict resolution goes more smoothlyMake discard_model_sampling True by defaultMade patch_hook_weight_to_device respect set_func and convert_funcSkip layer guidance now works on hydit model.Update README.md (#5707)set LTX min length to 1 for t2i (#5750)Skip layer guidance node now works on LTX-Video.Some improvements to the lowvram unloading.Fix some tiled VAE decoding issues with LTX-Video.v0.3.4v0.3.4LTXV lowvram fixes.Some fixes to the lowvram system.v0.3.3v0.3.3Merge branch 'master' into patch_hooks_improved_memoryRemove prints.v0.3.2v0.3.2Remove print.Merge branch 'master' into patch_hooks_improved_memoryAdd LTX-Video support to the Readme.Support Lightricks LTX-Video model.Fix mask issue in some attention functions.Merge branch 'master' into patch_hooks_improved_memoryRevert change, does not workAllow both FLOATS and FLOAT for floats_strength inputOn Cond/Cond Pair nodes, removed opt_ prefix from optional inputsSupport new flux model variants.v0.3.1v0.3.1Add some new weight patching functionality.Update readme.Turn off hook patch caching when only 1 hook present in sampling, replace some current_hook = None with calls to self.patch_hooks(None) instead to avoid a potential edge caseFix mask issue with attention_xformers.Skip layer guidance now works on stable audio model.v0.3.0v0.3.0