nodes/xiaden-comfyui-imgutils
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix README requirements: ComfyUI ships onnxruntime, nothing to replace

Browse Source
This commit is contained in:
xiaden
2026-07-03 01:25:38 +00:00
parent 1ff8c83add
commit 39dfd203fd
1 changed files with 2 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+2 -3
View File
@@ -86,9 +86,8 @@ Or install via ComfyUI-Manager (once published to the Registry).
## Requirements
- Python >= 3.10
- `dghs-imgutils >= 0.19.0` — the underlying analysis library
- `onnxruntime >= 1.17` — for ONNX model inference (auto-installed by `dghs-imgutils` if missing)
- **GPU recommended** — replace `onnxruntime` with `onnxruntime-gpu` for faster inference
- `dghs-imgutils >= 0.19.0` — the underlying analysis library (installed via `requirements.txt`)
- `onnxruntime` — for ONNX model inference. ComfyUI already ships both `onnxruntime` (CPU) and `onnxruntime-gpu` (CUDA) in its venv. The `dghs-imgutils[gpu]` extra also pulls `onnxruntime-gpu` on install, so there is nothing to manually install or replace.
Models are downloaded automatically from HuggingFace Hub on first use and cached to `~/.cache/huggingface/hub/`. No manual model downloads needed.