comfyanonymous
5da23d7f05
No need to check filename extensions to detect shuffle controlnet.
2023-08-28 16:49:06 -04:00
comfyanonymous
e4957ff97e
Put clip vision outputs on the CPU.
2023-08-28 16:26:11 -04:00
comfyanonymous
256eb57284
Load clipvision model to GPU for faster performance.
2023-08-28 15:29:27 -04:00
comfyanonymous
4fb6163a21
Text encoder should initially load on the offload_device not the regular.
2023-08-28 15:08:45 -04:00
comfyanonymous
201631e61d
Move ModelPatcher to model_patcher.py
2023-08-28 14:51:31 -04:00
comfyanonymous
daae7db069
Implement loras with norm keys.
2023-08-28 11:20:06 -04:00
comfyanonymous
ae3f7060d8
Enable bf16-vae by default on ampere and up.
2023-08-27 23:06:19 -04:00
comfyanonymous
6932eda1fb
Fallback to slice attention if xformers doesn't support the operation.
2023-08-27 22:24:42 -04:00
comfyanonymous
1b9a6a9599
Make --bf16-vae work on torch 2.0
2023-08-27 21:33:53 -04:00
comfyanonymous
90bfcef833
Fix lowvram model merging.
2023-08-26 11:52:07 -04:00
comfyanonymous
30d39b387d
The new smart memory management makes this unnecessary.
2023-08-25 18:02:15 -04:00
comfyanonymous
a2602fc4f9
Move controlnet code to comfy/controlnet.py
2023-08-25 17:33:04 -04:00
comfyanonymous
7fb8cbb5ac
Move lora code to comfy/lora.py
2023-08-25 17:11:51 -04:00
comfyanonymous
145e279e6c
Move text_projection to base clip model.
2023-08-24 23:43:48 -04:00
comfyanonymous
4731c0b618
Code cleanups.
2023-08-24 19:39:18 -04:00
comfyanonymous
74d1dfb0ad
Try to free enough vram for control lora inference.
2023-08-24 17:20:54 -04:00
comfyanonymous
5dbbb2c93c
Fix potential issue with text projection matrix multiplication.
2023-08-24 00:54:16 -04:00
comfyanonymous
e340ef7852
Always shift text encoder to GPU when the device supports fp16.
2023-08-23 21:45:00 -04:00
comfyanonymous
5ef57a983b
Even with forced fp16 the cpu device should never use it.
2023-08-23 21:38:28 -04:00
comfyanonymous
1aff0360c3
Initialize text encoder to target dtype.
2023-08-23 21:01:15 -04:00
comfyanonymous
e7fc7fb557
Save memory by storing text encoder weights in fp16 in most situations.
...
Do inference in fp32 to make sure quality stays the exact same.
2023-08-23 01:08:51 -04:00
comfyanonymous
ed16480867
All resolutions now work with t2i adapter for SDXL.
2023-08-22 16:23:54 -04:00
comfyanonymous
b168bdf3e5
T2I adapter SDXL.
2023-08-22 14:40:43 -04:00
comfyanonymous
08af73e450
Controlnet/t2iadapter cleanup.
2023-08-22 01:06:26 -04:00
comfyanonymous
f29b9306fd
Fix control lora not working in fp32.
2023-08-21 20:38:31 -04:00
comfyanonymous
b982fd039e
Fix ControlLora on lowvram.
2023-08-21 00:54:04 -04:00
comfyanonymous
819c4a42d3
Remove autocast from controlnet code.
2023-08-20 21:47:32 -04:00
comfyanonymous
37a6cb2649
Small cleanups.
2023-08-20 14:56:47 -04:00
Simon Lui
a670a3f848
Further tuning and fix mem_free_total.
2023-08-20 14:19:53 -04:00
Simon Lui
af8959c8a9
Add ipex optimize and other enhancements for Intel GPUs based on recent memory changes.
2023-08-20 14:19:51 -04:00
comfyanonymous
56901bd7c6
--disable-smart-memory now disables loading model directly to vram.
2023-08-20 04:00:53 -04:00
comfyanonymous
225a5f9f1f
Free more memory before VAE encode/decode.
2023-08-19 12:13:13 -04:00
comfyanonymous
01a6f9b116
Fix issue with gligen.
2023-08-18 16:32:23 -04:00
comfyanonymous
280659a6ee
Support for Control Loras.
...
Control loras are controlnets where some of the weights are stored in
"lora" format: an up and a down low rank matrice that when multiplied
together and added to the unet weight give the controlnet weight.
This allows a much smaller memory footprint depending on the rank of the
matrices.
These controlnets are used just like regular ones.
2023-08-18 11:59:51 -04:00
comfyanonymous
398390a76f
ReVision support: unclip nodes can now be used with SDXL.
2023-08-18 11:59:36 -04:00
comfyanonymous
e246c23708
Add support for clip g vision model to CLIPVisionLoader.
2023-08-18 11:13:29 -04:00
Alexopus
a5a8d25943
Fix referenced before assignment
...
For https://github.com/BlenderNeko/ComfyUI_TiledKSampler/issues/13
2023-08-17 22:30:07 +02:00
comfyanonymous
bcf55c1446
Fix issue with not freeing enough memory when sampling.
2023-08-17 15:59:56 -04:00
comfyanonymous
d8f9334347
Fix bug with lowvram and controlnet advanced node.
2023-08-17 13:38:51 -04:00
comfyanonymous
e075077ad8
Fix potential issues with patching models when saving checkpoints.
2023-08-17 11:07:08 -04:00
comfyanonymous
21e07337ed
Add --disable-smart-memory for those that want the old behaviour.
2023-08-17 03:12:37 -04:00
comfyanonymous
197ab43811
Fix issue with regular torch version.
2023-08-17 01:58:54 -04:00
comfyanonymous
a216b56591
Smarter memory management.
...
Try to keep models on the vram when possible.
Better lowvram mode for controlnets.
2023-08-17 01:06:34 -04:00
comfyanonymous
e4ffcf2c61
Support small diffusers controlnet so both types are now supported.
2023-08-16 12:45:56 -04:00
comfyanonymous
1a21a2271e
Support diffusers mini controlnets.
2023-08-16 12:28:01 -04:00
comfyanonymous
9e0c084148
Fix clip vision issue with old transformers versions.
2023-08-16 11:36:22 -04:00
comfyanonymous
dcbf839d22
Fix potential issue with batch size and clip vision.
2023-08-16 11:05:11 -04:00
comfyanonymous
601e4a9865
Refactor unclip code.
2023-08-14 23:48:47 -04:00
comfyanonymous
736e2e8f49
CLIPVisionEncode can now encode multiple images.
2023-08-14 16:54:05 -04:00
comfyanonymous
87f1037cf5
Remove 3m from PR #1213 because of some small issues.
2023-08-14 00:48:45 -04:00
