nodes/yuvraj108c-ComfyUI-Rife-Tensorrt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Improve cuda graph tooltip with RAM/resolution caveats

Browse Source 
Advises users to disable cuda graph if experiencing high RAM usage
or errors with variable input resolutions.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
reaperhammer
2026-05-09 18:39:17 +12:00
co-authored by Claude Opus 4.7
parent 203afb5f2a
commit 7cf8989cd3
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
__init__.py
+1 -1
View File
@@ -144,7 +144,7 @@ class RifeTensorrt:
"rife_trt_model": ("RIFE_TRT_MODEL", {"tooltip": "Tensorrt model built and loaded"}),
"clear_cache_after_n_frames": ("INT", {"default": 100, "min": 1, "max": 1000, "tooltip": "Clear CUDA cache after processing this many frames"}),
"multiplier": ("INT", {"default": 2, "min": 1, "tooltip": "Frame interpolation multiplier"}),
"use_cuda_graph": ("BOOLEAN", {"default": True, "tooltip": "Use CUDA graph for better performance"}),
"use_cuda_graph": ("BOOLEAN", {"default": True, "tooltip": "Use CUDA graph for better performance. Disable if experiencing high RAM usage or errors with variable input resolutions."}),
"keep_model_loaded": ("BOOLEAN", {"default": False, "tooltip": "Keep model loaded in memory after processing"}),
},
}