DataCTE / SDXL-Training-ImprovementsLinks
π Research-focused SDXL training framework exploring novel optimization approaches. Goals include enhanced image quality, training stability & comprehensive monitoring. β Performance-focused research framework.
β18Updated 4 months ago
Alternatives and similar repositories for SDXL-Training-Improvements
Users that are interested in SDXL-Training-Improvements are comparing it to the libraries listed below
Sorting:
- Vision Transformers Needs Registers. And Gated MLPs. And +20M params. Tiny modality gap ensues!β47Updated 4 months ago
- MoD Control Tile Upscaler for SDXL Pipelineβ61Updated 7 months ago
- Janky implementation of DiffuseHigh for ComfyUIβ36Updated 5 months ago
- β67Updated 4 months ago
- Gradio UI for training video models using finetrainersβ32Updated 5 months ago
- Inference code of "Golden Noise for Diffusion Models: A Learning Framework".β34Updated 10 months ago
- https://github.com/xie-lab-ml/Golden-Noise-for-Diffusion-Models for ComfyUIβ17Updated 10 months ago
- β19Updated last year
- A comprehensive codebase for training and finetuning Image <> Latent models.β45Updated 7 months ago
- ComfyUI unofficial implementation of Thera - Aliasing-Free Arbitrary-Scale Super-Resolution with Neural Heat Fieldsβ35Updated 5 months ago
- RadialAttention in ComfyUI native workflowβ72Updated last week
- β61Updated 5 months ago
- Adaptive ODE Solvers for ComfyUIβ52Updated last year
- Comfyui implementation of Regional Adaptive Sampling, for Flux and HunYuanVideoβ22Updated 2 weeks ago
- Create Latents with Perlin Noise in any shape (dimensionality). Works with Flux, SD3 and other 16d latent models.β34Updated last year
- Batched Runge-Kutta Samplers for ComfyUIβ61Updated last year
- β32Updated last year
- Experimental method to use reference video to drive motion in generations without training in ComfyUI.β36Updated last year
- Model code for inferencing T5β66Updated 7 months ago
- A custom node for ComfyUI that adds cinematic and movie scene styles to video generation prompts. This node helps create more dynamic andβ¦β45Updated 9 months ago
- β52Updated 7 months ago
- ComfyUI powertools for SD1.5 and SDXL model mergingβ92Updated 6 months ago
- A ComfyUI custom node that loads and applies B-LoRA models.β75Updated last year
- Allows to sample without generating any uncond with Stable Diffusion!β49Updated last year
- ComfyUI implementation of the paper https://arxiv.org/pdf/2410.02416 "APG".β33Updated 10 months ago
- Explore how Flux Dev responds when you change the strengths of layers in the model.β21Updated last year
- Somewhat janky implementation of Sonar Sampling (momentum based sampling) for ComfyUI along with an assortment of advanced noise tools (sβ¦β47Updated 2 months ago
- Wildly unsound and experimental sampling for ComfyUIβ28Updated 2 months ago
- Auto-MBW for ComfyUI loosely based on sdweb-auto-MBWβ15Updated last year
- Trajectory Consistency Distillation (TCD) sampling for ComfyUIβ84Updated last year