Heads up: statuses are automated heuristics from commit dates, registry metadata and issue titles. "Stale" can mean "finished and stable"; "Error spike" means error-titled issues are unusually high for that pack lately — worth checking recent issues, not a verdict that it is broken. Always check the repo.

Comfy-WaveSpeed Active

The all in one inference optimization solution for ComfyUI, universal, flexible, and fast.

Why: last commit 59 days ago

Repo	chengzeyi/Comfy-WaveSpeed (default branch main)
Registry	wavespeed · publisher chengzeyi · latest 1.1.8 (2025-01-20) · in Manager list: no
Popularity	rank #71 · 280,146 downloads · 1238 stars
Tags	Video Loaders, quantization & speed SDXL SD3 Flux LTX-Video
Star velocity	collecting (nightly snapshots started 2026-10-19)
Error issues	0 error-titled issues opened in last 30 days (open+closed) · 0.0 per 100K downloads · baseline 0.17/month over prior 6 months (1 total) · 0.0× baseline (Error spike needs ≥3, ≥2× baseline, ≥0.5 per 100K)
Last commit	2026-08-21 (59 days ago)
Open issues	92 issues, 5 PRs (fetched 2026-10-19)
pyproject.toml	present · version 1.1.8 · requires-comfyui — · frontend — · requires-python — · 0 deps
Registry ComfyUI req.

Error-titled issues, last 30 days

None

Open issues mentioning errors / breakage

#147 FBCache crashes on Flux 2 Dev — kwargs / y=None / txt_norm fixes (working diff inside) 2026-04-30
#146 when i use flux t2i workflow with wavespeed,error occurs:create_patch_flux_forward_orig.<locals>.forward_orig() got an unexpected keyword argument 'timestep_zer 2026-03-19
#144 Great loader. Caching and torch broken 2025-12-19
#131 Compile model + Apply first block cache doesn`t work together 2025-04-27
#125 flux WaveSpeed_controlnet .json not working with Canny effect? Request for support, thank you. 2025-03-26
#118 Error(s) in loading state_dict for T5: size mismatch for shared.weight: copying a param with shape torch.Size([256384, 4096]) from checkpoint, the shape in curr 2025-03-15
#96 Bypass on Compile model+ doesn't work 2025-01-25
#89 You can suppress this exception and fall back to eager by setting: import torch._dynamo torch._dynamo.config.suppress_errors = True 2025-01-19
#74 Error running SamplerCustomAdvanced 2025-01-14
#67 error when i use LTX with compile node, but complie node work with flux 2025-01-13
#53 Error running after update 2025-01-12
#43 Hunyuan Video error 2025-01-11
#42 ImportError: DLL load failed while importing cuda_utils: 找不到指定的模块。 2025-01-11
#37 3090 LTXV - SampleCustom Error 2025-01-11
#34 I was able to get torch to compile in Windows for Flux and LTX video, but HunyuanVideo compile fails. 2025-01-11

Recent commits

2026-08-21 [main] docs: refresh hosted links and add a hosted-inference section (#150)
2025-08-02 [main] make it work with latest comfyui
2025-03-27 [main] add links to wavespeed.ai
2025-03-22 [main] fix: Remove unsupported `modulation_dims_img` and `modulation_dims_txt` args from kwargs This resolves the compatibility issue with ComfyUI
2025-02-07 [main] use reshape(-1) instead of flatten() to avoid torch.compile bug