Enable true multi gpu capability in Comfy UI using XDiT XFuser and FSDP managed by Ray
☆476Nov 2, 2026Updated this week
Alternatives and similar repositories for raylight
Users that are interested in raylight are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- This custom_node for ComfyUI adds one-click "Virtual VRAM" for any UNet and CLIP loader as well MultiGPU integration in WanVideoWrapper, …☆993Sep 20, 2026Updated last month
- ☆21Sep 16, 2026Updated last month
- ComfyUI extension that enables multi-GPU processing locally, remotely and in the cloud☆648Updated this week
- This suite of nodes unlocks high-performance parallel processing in ComfyUI by utilizing **Model Replication**. Unlike standard offloadin…☆70Aug 13, 2026Updated 2 months ago
- qwen-image and wan2.2 wan2.1 bridge☆194Apr 18, 2026Updated 6 months ago
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- Some Comfyui custom nodes for wan2.1 VACE, attempt to implement VACE video generation/editing in a better way.☆141Oct 31, 2025Updated last year
- ☆2,128Updated this week
- Divide and Conquer Node Suite☆148Nov 17, 2025Updated 11 months ago
- Force any Model, VAE, or CLIP to any GPU or CPU—and keep it there (or don't).☆37Feb 6, 2026Updated 8 months ago
- ☆100Oct 2, 2026Updated last month
- ComfyUI Nodes for AsymFlow and pi-Flow☆192Sep 3, 2026Updated last month
- High quality training free inpaint for every stable diffusion model. Supports ComfyUI☆1,480Sep 28, 2026Updated last month
- RadialAttention in ComfyUI native workflow☆121Dec 19, 2025Updated 10 months ago
- An experimental node☆46Oct 28, 2025Updated last year
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Custom node to load models in INT8 for 1.5~2X Speed gains on 30 series cards.☆313Jun 26, 2026Updated 4 months ago
- ☆70Oct 7, 2025Updated last year
- FSampler is a training‑free, sampler‑agnostic acceleration layer for diffusion sampling.☆133Feb 28, 2026Updated 8 months ago
- ☆1,378Updated this week
- An upscaler node for flow-matching models like Qwen, applying the DemoFusion approach☆63Jan 29, 2026Updated 9 months ago
- Training-free steering dials for MiniMax H3: turn prompt pairs into a 14 KB slider (Freeze <-> Zoomies motion and more)☆30Updated this week
- ☆28Updated this week
- ComfyUI implemtation for NAG☆320Nov 3, 2025Updated 11 months ago
- ComfyUI's Photoshop-like layered canvas editor to your ComfyUI workflow. This node is perfect for complex compositing, inpainting, and ou…☆352Aug 21, 2026Updated 2 months ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- RAM is all you need☆284Sep 20, 2026Updated last month
- ComfyUI custom nodes providing unified image and mask resizing with support for multiple scaling modes, aspect-ratio preservation, center…☆214Sep 21, 2026Updated last month
- ☆67May 3, 2025Updated last year
- A ComfyUI node implementation for ByteDance's Sa2VA☆96Dec 22, 2025Updated 10 months ago
- MagicNodes, it's a plug-and-play multi-pass "render-machine" for SD/SDXL models. Simple one-node start, expert-grade results. Core is ZeR…☆92May 16, 2026Updated 5 months ago
- Some useful nodes for ComyUI☆47Oct 20, 2026Updated last week
- A repository of Python & PyTorch scripts which (currently) converts .safetensors models into scaled FP8 variants, utilizing gradient desc…☆26Aug 8, 2025Updated last year
- ComfyUI DyPE+SEGA, enabling artifact-free 4K+ image generation: Z-Image, Qwen, Flux, Krea2☆633Oct 24, 2026Updated last week
- Training-free Spectrum acceleration for ComfyUI’s native MiniMax H3 audio-video model. Uses Chebyshev ridge feature forecasting to skip s…☆713Updated this week
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- MiniMax-H3 multi-GPU parallel inference for ComfyUI | 多卡并行加速节点:2-8 GPU Ulysses sequence parallel, bit-identical video+audio generation☆38Sep 28, 2026Updated last month
- ComfyUI nodes and tested workflows for MiniMax H3 still-image generation, reference editing, and Qwen detail refinement.☆214Sep 30, 2026Updated last month
- PiD Decode Custom Node☆34May 25, 2026Updated 5 months ago
- [ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models☆3,961Sep 6, 2026Updated last month
- [CVPR 2026] Adaptive Spectral Feature Forecasting for Diffusion Sampling Acceleration☆139Sep 21, 2026Updated last month
- Click-to-refine + smart inpaint sampler node for ComfyUI (FLUX 2 Klein). One node: generate, then click / paint / drag to refine, or do l…☆141Aug 8, 2026Updated 2 months ago
- ComfyUI node that removes the 2px pixel grid left by the Qwen Image, Qwen Image 2.1 and Wan 2.1 VAEs (Krea2, Qwen Image, Anima). Auto-cal…☆97Updated this week