# Use-sage-attention BUG?

**URL:** <https://forum.comfy.org/t/use-sage-attention-bug/4583>\
**Category:** Custom Nodes\
**Created:** [August 24, 2026, 7:15am UTC](https://forum.comfy.org/t/use-sage-attention-bug/4583 "2026-08-24T07:15:02Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![thx](https://avatars.discourse-cdn.com/v4/letter/t/db5fbb/32.png) [@thx](https://forum.comfy.org/u/thx)\
**Post date:** [August 24, 2026, 7:15am UTC](https://forum.comfy.org/t/use-sage-attention-bug/4583/1 "2026-08-24T07:15:02Z")

</div>

C:\Users\Master\>set SAGEATTN=1

C:\Users\Master\>cd /d D:\Master\ComfyUI\_windows\_portable

D:\Master\ComfyUI\_windows\_portable\>.\python\_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --listen 0.0.0.0 --use-sage-attention  
[32m[INFO][0m setup plugin alembic.autogenerate.schemas  
[32m[INFO][0m setup plugin alembic.autogenerate.tables  
[32m[INFO][0m setup plugin alembic.autogenerate.types  
[32m[INFO][0m setup plugin alembic.autogenerate.constraints  
[32m[INFO][0m setup plugin alembic.autogenerate.defaults  
[32m[INFO][0m setup plugin alembic.autogenerate.comments  
[INFO] Found comfy\_kitchen backend triton: {‘available’: True, ‘disabled’: True, ‘unavailable\_reason’: None, ‘capabilities’: [‘adaln’, ‘apply\_rope’, ‘apply\_rope1’, ‘apply\_rope1\_’, ‘apply\_rope\_’, ‘apply\_rope\_split\_half’, ‘apply\_rope\_split\_half1’, ‘apply\_rope\_split\_half1\_’, ‘apply\_rope\_split\_half\_’, ‘dequantize\_nvfp4’, ‘dequantize\_per\_tensor\_fp8’, ‘int8\_linear’, ‘quantize\_and\_rotate\_rowwise’, ‘quantize\_int8\_rowwise’, ‘quantize\_mxfp8’, ‘quantize\_nvfp4’, ‘quantize\_per\_tensor\_fp8’, ‘rms\_adaln’, ‘rms\_rope’, ‘rms\_rope1’, ‘rms\_rope1\_’, ‘rms\_rope\_’, ‘rms\_rope\_split\_half’, ‘rms\_rope\_split\_half1’, ‘rms\_rope\_split\_half1\_’, ‘rms\_rope\_split\_half\_’]}  
[INFO] Found comfy\_kitchen backend eager: {‘available’: True, ‘disabled’: False, ‘unavailable\_reason’: None, ‘capabilities’: [‘adaln’, ‘apply\_rope’, ‘apply\_rope1’, ‘apply\_rope1\_’, ‘apply\_rope\_’, ‘apply\_rope\_split\_half’, ‘apply\_rope\_split\_half1’, ‘apply\_rope\_split\_half1\_’, ‘apply\_rope\_split\_half\_’, ‘convrot\_w4a4\_linear’, ‘dequantize\_convrot\_w4a4\_weight’, ‘dequantize\_int8\_convrot\_weight’, ‘dequantize\_int8\_convrot\_weight\_dtype’, ‘dequantize\_int8\_embedding’, ‘dequantize\_int8\_simple’, ‘dequantize\_int8\_simple\_dtype’, ‘dequantize\_mxfp8’, ‘dequantize\_nvfp4’, ‘dequantize\_per\_tensor\_fp8’, ‘gemv\_awq\_w4a16’, ‘int8\_linear’, ‘prepare\_int4\_weight\_for\_int8\_linear’, ‘quantize\_and\_rotate\_rowwise’, ‘quantize\_convrot\_w4a4\_weight’, ‘quantize\_int8\_convrot\_weight’, ‘quantize\_int8\_rowwise’, ‘quantize\_int8\_tensorwise’, ‘quantize\_mxfp8’, ‘quantize\_nvfp4’, ‘quantize\_per\_tensor\_fp8’, ‘quantize\_svdquant\_w4a4’, ‘rms\_adaln’, ‘rms\_rope’, ‘rms\_rope1’, ‘rms\_rope1\_’, ‘rms\_rope\_’, ‘rms\_rope\_split\_half’, ‘rms\_rope\_split\_half1’, ‘rms\_rope\_split\_half1\_’, ‘rms\_rope\_split\_half\_’, ‘scaled\_mm\_mxfp8’, ‘scaled\_mm\_nvfp4’, ‘scaled\_mm\_svdquant\_w4a4’, ‘stochastic\_rounding\_fp8’]}  
[INFO] Found comfy\_kitchen backend hip: {‘available’: False, ‘disabled’: False, ‘unavailable\_reason’: ‘PyTorch ROCm/HIP runtime not available’, ‘capabilities’: []}  
[INFO] Found comfy\_kitchen backend cuda: {‘available’: True, ‘disabled’: False, ‘unavailable\_reason’: None, ‘capabilities’: [‘adaln’, ‘apply\_rope’, ‘apply\_rope1’, ‘apply\_rope1\_’, ‘apply\_rope\_’, ‘apply\_rope\_split\_half’, ‘apply\_rope\_split\_half1’, ‘apply\_rope\_split\_half1\_’, ‘apply\_rope\_split\_half\_’, ‘convrot\_w4a4\_linear’, ‘dequantize\_convrot\_w4a4\_weight’, ‘dequantize\_int8\_convrot\_weight’, ‘dequantize\_int8\_convrot\_weight\_dtype’, ‘dequantize\_int8\_simple’, ‘dequantize\_int8\_simple\_dtype’, ‘dequantize\_nvfp4’, ‘dequantize\_per\_tensor\_fp8’, ‘gemv\_awq\_w4a16’, ‘int8\_linear’, ‘prepare\_int4\_weight\_for\_int8\_linear’, ‘quantize\_and\_rotate\_rowwise’, ‘quantize\_convrot\_w4a4\_weight’, ‘quantize\_int8\_convrot\_weight’, ‘quantize\_int8\_rowwise’, ‘quantize\_int8\_tensorwise’, ‘quantize\_mxfp8’, ‘quantize\_nvfp4’, ‘quantize\_per\_tensor\_fp8’, ‘quantize\_svdquant\_w4a4’, ‘rms\_adaln’, ‘rms\_rope’, ‘rms\_rope1’, ‘rms\_rope1\_’, ‘rms\_rope\_’, ‘rms\_rope\_split\_half’, ‘rms\_rope\_split\_half1’, ‘rms\_rope\_split\_half1\_’, ‘rms\_rope\_split\_half\_’, ‘scaled\_mm\_nvfp4’, ‘scaled\_mm\_svdquant\_w4a4’, ‘stochastic\_rounding\_fp8’]}  
[INFO] Checkpoint files will always be loaded safely.  
[INFO] Total VRAM 49139 MB, total RAM 97948 MB  
[INFO] pytorch version: 2.13.0+cu130  
[INFO] Set vram state to: NORMAL\_VRAM  
[INFO] Device: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync  
[INFO] Using async weight offloading with 2 streams  
[INFO] Enabled pinned memory 39179.0  
[INFO] Using sage attention  
aimdo: src-win/cuda-detour.c:38:INFO:aimdo\_setup\_hooks: installing 6 hooks  
aimdo: src-win/cuda-detour.c:28:DEBUG:install\_hook\_entries: hooks successfully installed  
aimdo: src/control.c:262:INFO:comfy-aimdo NVML pressure enabled  
aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 4090 runtime\_luid=00000000:0001784b dxgi\_luid=00000000:0001784b  
aimdo: src/control.c:277:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 4090 (VRAM: 49139 MB)  
[INFO] DynamicVRAM support detected and enabled

INFO\] Python version: 3.13.14 (tags/v3.13.14:fd17997, Jun 10 2026, 13:03:48) \[MSC v.1944 64 bit (AMD64)

[INFO] ComfyUI version: 0.30.0  
[INFO] comfy-aimdo version: 0.4.11  
[INFO] comfy-kitchen version: 0.2.26  
[INFO] comfyui-frontend-package version: 1.47.11  
[INFO] comfyui-workflow-templates version: 0.11.27  
[INFO] comfyui-embedded-docs version: 0.5.9  
[INFO] comfy-kitchen version: 0.2.26  
[INFO] comfy-aimdo version: 0.4.11  
[INFO] [Prompt Server] web root: D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\comfyui\_frontend\_package\static  
[INFO] Asset seeder disabled  
[INFO] No OpenGL\_accelerate module loaded: No module named ‘OpenGL\_accelerate’

INFO

Import times for custom nodes:  
[INFO] 0.0 seconds: D:\Master\ComfyUI\_windows\_portable\ComfyUI\custom\_nodes\websocket\_image\_save.py

INFO

[INFO] Context impl SQLiteImpl.  
[INFO] Will assume non-transactional DDL.  
[INFO] Using RAM pressure cache.  
[INFO] Starting server

[INFO] To see the GUI go to: [http://0.0.0.0:8188](http://0.0.0.0:8188)  
[INFO] got prompt  
[ERROR] Error handling request from 127.0.0.1  
Traceback (most recent call last):  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_protocol.py”, line 575, in \_handle\_request  
resp = await request\_handler(request)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_app.py”, line 559, in \_handle  
return await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_middlewares.py”, line 117, in impl  
return await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\middleware\cache\_middleware.py”, line 27, in cache\_control  
response: web.Response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 98, in deprecation\_warning  
response: web.Response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 193, in origin\_only\_middleware  
response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 466, in upload\_image  
post = await request.post()  
^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_request.py”, line 774, in post  
while chunk := await field.read\_chunk(size=DEFAULT\_CHUNK\_SIZE):  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\multipart.py”, line 361, in read\_chunk  
chunk = await self.\_read\_chunk\_from\_stream(size)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\multipart.py”, line 420, in \_read\_chunk\_from\_stream  
chunk += await self.\_content.read(size)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\streams.py”, line 413, in read  
raise self.\_exception  
ConnectionResetError: Connection lost  
[ERROR] Error handling request from 127.0.0.1  
Traceback (most recent call last):  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_protocol.py”, line 575, in \_handle\_request  
resp = await request\_handler(request)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_app.py”, line 559, in \_handle  
return await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_middlewares.py”, line 117, in impl  
return await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\middleware\cache\_middleware.py”, line 27, in cache\_control  
response: web.Response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 98, in deprecation\_warning  
response: web.Response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 193, in origin\_only\_middleware  
response = await handler(request)  
^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\server.py”, line 466, in upload\_image  
post = await request.post()  
^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\web\_request.py”, line 774, in post  
while chunk := await field.read\_chunk(size=DEFAULT\_CHUNK\_SIZE):  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\multipart.py”, line 361, in read\_chunk  
chunk = await self.\_read\_chunk\_from\_stream(size)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\multipart.py”, line 420, in \_read\_chunk\_from\_stream  
chunk += await self.\_content.read(size)  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\aiohttp\streams.py”, line 413, in read  
raise self.\_exception  
ConnectionResetError: Connection lost  
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float32  
[INFO] got prompt  
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float16  
[INFO] Found quantization metadata version 1  
[INFO] Using MixedPrecisionOps for text encoder  
[INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16  
[INFO] Requested to load MiniMaxH3VideoVAE  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Requested to load MiniMaxH3TEModel\_  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] Found quantization metadata version 1  
[INFO] Detected mixed precision quantization  
[INFO] Using mixed precision operations  
[INFO] Native ops: int8\_tensorwise, convrot\_w4a4, float8\_e4m3fn, float8\_e5m2 , emulated ops: nvfp4, mxfp8  
[INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16  
[INFO] model\_type FLOW  
[INFO] Requested to load MiniMaxH3  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [01:05\<00:00, 3.27s/it]  
[INFO] Requested to load MiniMaxH3AudioVAE  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 106.91 seconds  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [01:06\<00:00, 3.30s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 76.53 seconds  
[INFO] got prompt  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
0%| | 0/20 [00:00\<?, ?it/s, Model Initializing …]100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [19:01\<00:00, 57.10s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 00:20:23  
[INFO] got prompt  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] got prompt  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [19:25\<00:00, 58.26s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 00:21:01  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [17:52\<00:00, 53.63s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 00:19:23  
[INFO] got prompt  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [17:59\<00:00, 53.99s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 00:19:36  
[INFO] got prompt  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
60%|███████████████████████████████████████████████████████████████████▏ | 12/20 [11:22\<07:35, 56.88s/it][INFO] got prompt  
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 20/20 [18:57\<00:00, 56.88s/it]  
[INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Prompt executed in 00:20:31  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB.  
[INFO] Model MiniMaxH3TEModel\_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB.  
[INFO] 0 models unloaded.  
[INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 0 patches attached. Force pre-loaded 210 weights: 1175 KB.  
0%| | 0/20 [00:00\<?, ?it/s, Model Initializing …][ERROR] Error running sage attention: CUDA error: an illegal memory access was encountered  
Search for `cudaErrorIllegalAddress’ in [CUDA Runtime API :: CUDA Toolkit Documentation](https://docs.nvidia.com/cuda/cuda-runtime-api/group __CUDART__ TYPES.html) for more information.  
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.  
For debugging consider passing CUDA\_LAUNCH\_BLOCKING=1  
, using pytorch attention instead.  
Fatal Python error: Aborted

Stack (most recent call first):  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\modules\attention.py”, line 581 in attention\_sage  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\modules\attention.py”, line 160 in wrapper  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\minimax\model.py”, line 181 in forward  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1789 in \_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1778 in \_wrapped\_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\minimax\model.py”, line 270 in forward  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1789 in \_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1778 in \_wrapped\_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\minimax\model.py”, line 629 in \_forward  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\ldm\minimax\model.py”, line 499 in forward  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1789 in \_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\nn\modules\module.py”, line 1778 in \_wrapped\_call\_impl  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\model\_base.py”, line 240 in \_apply\_model  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\model\_base.py”, line 196 in apply\_model  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 335 in \_calc\_cond\_batch  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 219 in \_calc\_cond\_batch\_outer  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 211 in calc\_cond\_batch  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 620 in sampling\_function  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1218 in predict\_noise  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1215 in outer\_predict\_noise  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1208 in \__call_\_  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 640 in \__call_\_  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\k\_diffusion\sampling.py”, line 1418 in res\_multistep  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\utils\_contextlib.py”, line 124 in decorate\_context  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\k\_diffusion\sampling.py”, line 1460 in sample\_res\_multistep  
File “D:\Master\ComfyUI\_windows\_portable\python\_embeded\Lib\site-packages\torch\utils\_contextlib.py”, line 124 in decorate\_context  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1005 in sample  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1235 in inner\_sample  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1260 in outer\_sample  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\patcher\_extension.py”, line 113 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\samplers.py”, line 1333 in sample  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\_extras\nodes\_custom\_sampler.py”, line 1049 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\_api\latest\_io.py”, line 1935 in EXECUTE\_NORMALIZED  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\comfy\_api\internal\__init_\_.py”, line 149 in wrapped\_func  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 306 in process\_inputs  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 318 in \_async\_map\_node\_over\_list  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 344 in get\_output\_data  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 545 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 789 in execute\_async  
File “asyncio\events.py”, line 89 in \_run  
File “asyncio\base\_events.py”, line 2060 in \_run\_once  
File “asyncio\base\_events.py”, line 683 in run\_forever  
File “asyncio\base\_events.py”, line 712 in run\_until\_complete  
File “asyncio\runners.py”, line 119 in run  
File “asyncio\runners.py”, line 196 in run  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\execution.py”, line 728 in execute  
File “D:\Master\ComfyUI\_windows\_portable\ComfyUI\main.py”, line 372 in prompt\_worker  
File “threading.py”, line 995 in run  
File “threading.py”, line 1044 in \_bootstrap\_inner  
File “threading.py”, line 1015 in \_bootstrap

Extension modules: sqlalchemy.cyextension.collections, sqlalchemy.cyextension.immutabledict, sqlalchemy.cyextension.processors, sqlalchemy.cyextension.resultproxy, sqlalchemy.cyextension.util, greenlet.\_greenlet, markupsafe.\_speedups, yaml.\_yaml, PIL.\_imaging, numpy.\_core.\_multiarray\_umath, numpy.linalg.\_umath\_linalg, torch.\_C, torch.\_C.\_dynamo.autograd\_compiler, torch.\_C.\_dynamo.eval\_frame, torch.\_C.\_dynamo.guards, torch.\_C.\_dynamo.utils, torch.\_C.\_fft, torch.\_C.\_linalg, torch.\_C.\_nested, torch.\_C.\_nn, torch.\_C.\_sparse, torch.\_C.\_special, numpy.random.\_common, numpy.random.bit\_generator, numpy.random.\_bounded\_integers, numpy.random.\_pcg64, numpy.random.\_generator, numpy.random.\_mt19937, numpy.random.\_philox, numpy.random.\_sfc64, numpy.random.mtrand, psutil.\_psutil\_windows, PIL.\_imagingft, \_cyutility, scipy.\_cyutility, scipy.\_lib.\_ccallback\_c, charset\_normalizer.md, charset\_normalizer.cd, scipy.ndimage.\_nd\_image, scipy.ndimage.\_rank\_filter\_1d, scipy.special.\_ufuncs\_cxx, scipy.linalg.\_fblas, scipy.linalg.\_flapack, scipy.linalg.\_batched\_linalg, scipy.linalg.\_cythonized\_array\_utils, scipy.linalg.\_solve\_toeplitz, scipy.linalg.\_internal\_matfuncs, scipy.linalg.\_linalg\_pythran, scipy.linalg.cython\_blas, scipy.linalg.cython\_lapack, scipy.linalg.\_decomp\_update, scipy.special.\_ellip\_harm\_2, scipy.special.\_special\_ufuncs, scipy.special.\_gufuncs, scipy.special.\_ufuncs, scipy.special.\_specfun, scipy.special.\_comb, \_ni\_label, scipy.ndimage.\_ni\_label, regex.\_regex, scipy.integrate.\_odepack, scipy.integrate.\_quadpack, scipy.integrate.\_vode, scipy.integrate.\_dop, scipy.sparse.\_sparsetools, \_csparsetools, scipy.sparse.\_csparsetools, scipy.sparse.linalg.\_dsolve.\_superlu, scipy.sparse.linalg.\_eigen.arpack.\_arpacklib, scipy.sparse.linalg.\_propack, scipy.optimize.\_group\_columns, scipy.\_lib.messagestream, scipy.optimize.\_trlib.\_trlib, scipy.optimize.\_lbfgsb, \_moduleTNC, scipy.optimize.\_moduleTNC, scipy.optimize.\_slsqplib, scipy.optimize.\_minpack, scipy.optimize.\_lsq.givens\_elimination, scipy.optimize.\_zeros, scipy.\_lib.\_uarray.\_uarray, scipy.linalg.\_decomp\_interpolative, scipy.optimize.\_bglu\_dense, scipy.optimize.\_lsap, scipy.spatial.\_ckdtree, scipy.spatial.\_hausdorff, scipy.spatial.\_distance\_wrap, scipy.spatial.\_qhull, scipy.spatial.\_voronoi, scipy.spatial.transform.\_rotation\_cy, scipy.spatial.transform.\_rigid\_transform\_cy, scipy.optimize.\_direct, scipy.interpolate.\_fitpack, scipy.interpolate.\_dierckx, scipy.interpolate.\_ppoly, scipy.interpolate.\_interpnd, scipy.interpolate.\_rbfinterp\_pythran, scipy.interpolate.\_rgi\_cython, scipy.special.cython\_special, scipy.stats.\_stats, scipy.stats.\_biasedurn, scipy.stats.\_levy\_stable.levyst, scipy.stats.\_stats\_pythran, scipy.stats.\_ansari\_swilk\_statistics, scipy.sparse.csgraph.\_tools, scipy.sparse.csgraph.\_shortest\_path, scipy.sparse.csgraph.\_traversal, scipy.sparse.csgraph.\_min\_spanning\_tree, scipy.sparse.csgraph.\_flow, scipy.sparse.csgraph.\_matching, scipy.sparse.csgraph.\_reordering, scipy.stats.\_sobol, scipy.stats.\_qmc\_cy, scipy.stats.\_rcont.rcont, scipy.stats.\_qmvnt\_cy, av.\_core, av.logging, av.buffer, av.audio.format, av.error, av.dictionary, av.container.pyio, av.format, av.index, av.utils, av.stream, av.container.streams, av.sidedata.encparams, av.sidedata.motionvectors, av.sidedata.sidedata, av.opaque, av.packet, av.container.input, av.bitstream, av.container.output, av.container.core, av.codec.context, av.video.format, av.video.reformatter, av.plane, av.video.plane, av.video.frame, av.video.stream, av.codec.hwaccel, av.codec.codec, av.frame, av.audio.layout, av.audio.plane, av.audio.frame, av.audio.stream, av.filter.link, av.filter.context, av.filter.graph, av.filter.filter, av.filter.loudnorm, av.audio.resampler, av.audio.codeccontext, av.audio.fifo, av.device, av.video.codeccontext, av.subtitles.stream, multidict.\_multidict, yarl.\_quoting\_c, propcache.\_helpers\_c, aiohttp.\_http\_writer, aiohttp.\_http\_parser, aiohttp.\_websocket.mask, aiohttp.\_websocket.reader\_c, frozenlist.\_frozenlist, requests.packages.charset\_normalizer.md, requests.packages.chardet.md, requests.packages.charset\_normalizer.cd, requests.packages.chardet.cd, scipy.signal.\_sigtools, scipy.signal.\_max\_len\_seq\_inner, scipy.signal.\_upfirdn\_apply, scipy.signal.\_spline, scipy.signal.\_sosfilt, scipy.signal.\_peak\_finding\_utils, cuda\_utils (total: 180)

D:\Master\ComfyUI\_windows\_portable\>set SAGEATTN=1

D:\Master\ComfyUI\_windows\_portable\>cd /d D:\Master\ComfyUI\_windows\_portable

D:\Master\ComfyUI\_windows\_portable\>.\python\_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --listen 0.0.0.0 --use-sage-attentionset SAGEATTN=1  
usage: main.py [-h] [–listen [IP]] [–port PORT] [–tls-keyfile TLS\_KEYFILE] [–tls-certfile TLS\_CERTFILE] [–enable-cors-header [ORIGIN]]  
[–max-upload-size MAX\_UPLOAD\_SIZE] [–base-directory BASE\_DIRECTORY] [–extra-model-paths-config PATH [PATH …]]  
[–output-directory OUTPUT\_DIRECTORY] [–temp-directory TEMP\_DIRECTORY] [–input-directory INPUT\_DIRECTORY] [–auto-launch]  
[–disable-auto-launch] [–cuda-device DEVICE\_ID] [–default-device DEFAULT\_DEVICE\_ID] [–cuda-malloc | --disable-cuda-malloc]  
[–force-fp32 | --force-fp16] [–fp32-unet | --fp64-unet | --bf16-unet | --fp16-unet | --fp8\_e4m3fn-unet | --fp8\_e5m2-unet |  
–fp8\_e8m0fnu-unet] [–fp16-vae | --fp32-vae | --bf16-vae] [–cpu-vae] [–fp8\_e4m3fn-text-enc | --fp8\_e5m2-text-enc |  
–fp16-text-enc | --fp32-text-enc | --bf16-text-enc] [–fp16-intermediates] [–force-channels-last] [–directml [DIRECTML\_DEVICE]]  
[–oneapi-device-selector SELECTOR\_STRING] [–supports-fp8-compute] [–enable-triton-backend] [–disable-triton-backend]  
[–preview-method [none,auto,latent2rgb,taesd]] [–preview-size PREVIEW\_SIZE] [–cache-ram [GB …] | --cache-classic |  
–cache-lru CACHE\_LRU | --cache-none | --high-ram] [–use-split-cross-attention | --use-quad-cross-attention |  
–use-pytorch-cross-attention | --use-sage-attention | --use-flash-attention] [–disable-xformers] [–force-upcast-attention |  
–dont-upcast-attention] [–enable-manager] [–disable-manager-ui | --enable-manager-legacy-ui] [–gpu-only | --highvram |  
–lowvram | --novram | --cpu] [–reserve-vram RESERVE\_VRAM] [–vram-headroom VRAM\_HEADROOM] [–disable-nvml-pressure]  
[–async-offload [NUM\_STREAMS]] [–disable-async-offload] [–disable-dynamic-vram] [–enable-dynamic-vram] [–fast-disk]  
[–force-non-blocking] [–default-hashing-function {md5,sha1,sha256,sha512}] [–disable-smart-memory] [–deterministic]  
[–fast [FAST …]] [–debug-hang] [–disable-pinned-memory] [–mmap-torch-files] [–disable-mmap] [–dont-print-server]  
[–quick-test-for-ci] [–windows-standalone-build] [–disable-metadata] [–disable-all-custom-nodes]  
[–whitelist-custom-nodes WHITELIST\_CUSTOM\_NODES [WHITELIST\_CUSTOM\_NODES …]] [–disable-api-nodes] [–multi-user]  
[–verbose [LEVEL FILE …]] [–log-stdout] [–front-end-version FRONT\_END\_VERSION] [–front-end-root FRONT\_END\_ROOT]  
[–user-directory USER\_DIRECTORY] [–models-directory MODELS\_DIRECTORY] [–enable-compress-response-body]  
[–comfy-api-base COMFY\_API\_BASE] [–database-url DATABASE\_URL] [–enable-assets] [–enable-asset-hashing]  
[–feature-flag KEY[=VALUE]] [–list-feature-flags]  
main.py: error: unrecognized arguments: --use-sage-attentionset SAGEATTN=1

D:\Master\ComfyUI\_windows\_portable\>cd /d D:\Master\ComfyUI\_windows\_portable

D:\Master\ComfyUI\_windows\_portable\>.\python\_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --listen 0.0.0.0 --use-sage-attention

---

<div class="post-metadata">

**Author:** ![Detrolio](https://avatars.discourse-cdn.com/v4/letter/d/3da27b/32.png) [@Detrolio](https://forum.comfy.org/u/Detrolio)\
**Post date:** [August 26, 2026, 6:07am UTC](https://forum.comfy.org/t/use-sage-attention-bug/4583/2 "2026-08-26T06:07:47Z")

</div>

I solved my install problems with google AI. Just search for „comfyui use sage attention“, the AI part will guide you through the installation parts.

---

<div class="post-metadata">

**Author:** ![thx](https://avatars.discourse-cdn.com/v4/letter/t/db5fbb/32.png) [@thx](https://forum.comfy.org/u/thx)\
**Post date:** [August 26, 2026, 6:30am UTC](https://forum.comfy.org/t/use-sage-attention-bug/4583/3 "2026-08-26T06:30:15Z")

</div>

Do you have any records on this? Could you send me a copy?

---

<div class="post-metadata">

**Author:** ![Detrolio](https://avatars.discourse-cdn.com/v4/letter/d/3da27b/32.png) [@Detrolio](https://forum.comfy.org/u/Detrolio)\
**Post date:** [August 26, 2026, 7:36am UTC](https://forum.comfy.org/t/use-sage-attention-bug/4583/4 "2026-08-26T07:36:53Z")

</div>

No, sry.