#132 · Primary category： Inference & Local Deploy
Comfy-WaveSpeed
https://wavespeed.ai/ [WIP] The all in one inference optimization solution for ComfyUI, universal, flexible, and fast.
Project last updated：08/21/26
GitHub Stars
1.2K
Forks
68
Contributors
4
License
MIT
Why we included this project
ComfyUI users who generate with FLUX, SDXL, or video models tend to spend most of their time staring at progress bars. Comfy-WaveSpeed targets exactly that wait time with a first-block caching scheme that reuses transformer computation when consecutive denoising steps produce similar results, plus an enhanced torch.compile wrapper that works with LoRAs. The pack ships ready-made workflows with suggested cache thresholds for models like FLUX, LTXV, HunyuanVideo, SD3.5, and SDXL, so you are not guessing at parameters, and typical speedups land in the 1.5x to 3x range. The caching node also runs standalone on machines where torch.compile is not an option, which makes it an easy first optimization to try. It is still labeled work-in-progress, so expect rough edges and frequent updates as more features land.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
ollama
Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
llama.cpp
LLM inference in C/C++
vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
gpt4all
GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.
LocalAI
LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.