huchukato
de53234631
Auto-disable CUDA graph in cloud environments - detect RunPod, Colab, Docker and disable by default for stability
2026-02-27 02:37:13 +01:00
huchukato
ccd3f3bf9c
Add CUDA graph fallback for persistent ERROR 35 in cloud environments - disable graph optimization and retry
2026-02-27 02:35:08 +01:00
huchukato
0c0f4b2d10
Fix TensorRT tensor address binding - add explicit set_tensor_address for all tensors before inference
2026-02-27 02:32:38 +01:00
huchukato
c19731ec58
Fix AttributeError by initializing inputs and outputs attributes in Engine constructor
2026-02-27 02:30:44 +01:00
huchukato
5ab7eedbb9
Add CUDA ERROR 35 recovery mechanism with retry and context reset for inference failures
2026-02-27 02:29:00 +01:00
huchukato
5aec40e052
Add comprehensive CUDA diagnostics and troubleshooting for TensorRT build failures
2026-02-27 02:15:50 +01:00
huchukato
08cb64243f
README Update
2026-02-21 09:32:40 +01:00
huchukato
ab8373d6c2
Replace text links with shield.io language badges for better visual consistency
2026-02-21 09:19:41 +01:00
huchukato
b8b2932de6
Add Italian README with badges and cross-link to main README
2026-02-21 09:17:41 +01:00
huchukato
24d5aa80af
Clarify legacy requirements_cu12.txt usage and improve installation documentation
2026-02-21 09:16:45 +01:00
huchukato
3cdf12ab66
Implement fully automatic CUDA detection and TensorRT installation on node load
2026-02-21 09:14:05 +01:00
huchukato
6eb23462d9
Add auto-install scripts for CUDA version detection and requirements installation
2026-02-21 09:11:35 +01:00
Chuck
b55f69970d
Add large resolution profile (720-1920px) for high-resolution content support
2026-02-04 03:55:20 +01:00
Chuck
ea738af937
Update TensorRT to latest version 10.15.1.29 for better compatibility
2026-02-04 03:41:51 +01:00
Chuck
343e25ce6c
Fix tensorrt_cu13_bindings version to use available .post1 release
2026-02-04 03:39:13 +01:00
silveroxides
46d7a50445
update default resolutions and bump version to 1.1.1
2026-01-10 00:14:36 +01:00
Silver
7c826189a6
Bump version from 1.0.0 to 1.1.0
2026-01-09 22:47:24 +01:00
Silver
1139843f55
Merge pull request #2 from silveroxides/fixes-alot
...
Fixes alot
2026-01-09 22:46:53 +01:00
silveroxides
5cff7f7dc7
feat: CUDA 13 default, auto-detection, and resolution profiles
...
Requirements updates:
- Default to CUDA 13 with TensorRT 10.14.1.48 (requirements.txt)
- Add cuda-toolkit version constraints (>=13.0.0,<13.1.0)
- Rename requirements_cu13.txt -> requirements_cu12.txt for legacy support
- Add cuda-toolkit constraint for CUDA 12 (>=12.8.0,<13.0.0)
CUDA toolkit detection:
- Add _setup_cuda_dll_path() to auto-detect CUDA installation
- Check CUDA_PATH, CUDA_HOME, and default Windows install paths
- Handle CUDA 13+ bin/x64 subdirectory for cudart64.dll
- Display download link if CUDA not found
Engine building updates:
- Add resolution profiles (small: 480-896px, medium: 720-1280px)
- Add 'custom' profile with CustomResolutionConfig node
- Include profile name in engine filename for separate caching
- Reduce default VRAM usage from ~15GB to ~2-3GB
2026-01-09 22:44:36 +01:00
silveroxides
8c4f68b5f3
try new requirements setup
2026-01-04 09:21:46 +01:00
Silver
040846632a
Disable ProgressBar because of performance issues
2025-12-16 18:38:20 +01:00
Silver
ee5daaddfa
Update pyproject.toml
2025-11-29 20:32:29 +01:00
Silver
45f84019b7
Update __init__.py
2025-11-29 20:31:14 +01:00
Silver
b62f2feffa
Update pyproject.toml
2025-11-29 20:26:53 +01:00
Silver
28e79bb3a2
Update publish_action.yml
2025-11-29 20:25:23 +01:00
silveroxides
115386f898
Add ComfyUI Registry Publishing
2025-11-29 20:24:33 +01:00
Silver
00cbebe151
Merge pull request #1 from silveroxides/PR14-reaperhammer
...
Merge pr from reaperhammer
2025-11-29 20:12:43 +01:00
silveroxides
d61e185923
trim trailing whitespace
2025-11-29 20:10:04 +01:00
reaperhammer
6cd119d9e5
fix memory issues with graphs when keep_model_loaded = true
2025-09-25 08:16:02 +12:00
reaperhammer
3e898b65f7
fix engine reload on batch jobs when keep_model_loaded = false
2025-09-25 07:54:13 +12:00
reaperhammer
16994ac3cb
Update readme.md
...
more readme updates
2025-09-25 00:42:28 +12:00
reaperhammer
8a2a1ae405
Update readme.md
...
Update the Python and Cuda version images as it works with these later versions that I'm using
2025-09-25 00:36:16 +12:00
reaperhammer
25d6c9f1d2
Make the node auto download/build models and update readme.md accordingly
2025-09-25 00:08:26 +12:00
reaperhammer
90b7af76d0
Merge remote-tracking branch 'refs/remotes/origin/master'
...
merge local branch with remote
2025-09-25 00:04:16 +12:00
reaperhammer
e0d17b55e7
Merge pull request #13 from ritlo/master
...
fix: add export_trt.py to sys.path to fix it for use with comfyui's python_embeded/python.exe
2025-09-24 23:55:35 +12:00
Yuvraj Seegolam
41686c5c64
Merge pull request #13 from ritlo/master
...
fix: add export_trt.py to sys.path to fix it for use with comfyui's python_embeded/python.exe
2025-09-23 11:06:13 +04:00
Ritvick Lohani
261efd6e5b
fix: add export_trt to sys.path to fix it for use with comfyui's python_embedded/python.exe
2025-09-23 02:24:44 +05:30
Yuvraj Seegolam
06b847dfa7
Merge pull request #12 from RioShiina47/master
...
fix cudart
2025-09-18 18:47:25 +04:00
RioShiina47
e971cfb71a
fix cudart
2025-09-16 20:44:51 +08:00
Yuvraj Seegolam
addc5be46d
add node img
2024-10-04 14:23:26 +04:00
yuvraj108c
08fa66cd1e
log new frames count
2024-10-04 13:36:16 +04:00
yuvraj108c
987334d48d
fix cuda graph
2024-10-04 13:32:57 +04:00
yuvraj108c
7491b548d1
cuda grapsh
2024-10-04 13:30:05 +04:00
yuvraj108c
116ce22440
keep_model_loaded
2024-10-03 23:11:13 +04:00
yuvraj108c
333d068dc1
add license
2024-10-03 14:20:09 +04:00
Yuvraj Seegolam
560f7752b1
add benchmarks
2024-10-03 14:09:52 +04:00
yuvraj108c
f522f38bb8
readme
2024-10-03 08:13:57 +00:00
yuvraj108c
37c7b9df6a
readme
2024-10-03 08:09:08 +00:00
yuvraj108c
c6aed0e8a5
cleanup
2024-10-03 07:17:18 +00:00
yuvraj108c
4952023d35
v1
2024-10-03 07:14:29 +00:00
huchukato