Commits · ae8754365356a56e9d97bd254ad756660b615b47 · chenpangpang / ComfyUI

23 Sep, 2023 1 commit
- Allow Intel GPUs to LoRA cast on GPU since it supports BF16 natively. · eec449ca
  Simon Lui authored Sep 22, 2023
  
  eec449ca
20 Sep, 2023 1 commit
- Only do the cast on the device if the device supports it. · 1cdfb3db
  comfyanonymous authored Sep 20, 2023
  
  1cdfb3db
17 Sep, 2023 1 commit
- Enable pytorch attention by default on xpu. · 321c5fa2
  comfyanonymous authored Sep 17, 2023
  
  321c5fa2
14 Sep, 2023 1 commit
- Don't run text encoders on xpu because there are issues. · 0966d3ce
  comfyanonymous authored Sep 14, 2023
  
  0966d3ce
04 Sep, 2023 1 commit
- Add a force argument to soft_empty_cache to force a cache empty. · 1938f5c5
  comfyanonymous authored Sep 04, 2023
  
  1938f5c5
03 Sep, 2023 1 commit
- Some fixes to generalize CUDA specific functionality to Intel or other GPUs. · 4a0c4ce4
  Simon Lui authored Sep 02, 2023
  
  4a0c4ce4
28 Aug, 2023 1 commit
- Enable bf16-vae by default on ampere and up. · b8c7c770
  comfyanonymous authored Aug 27, 2023
  
  b8c7c770
26 Aug, 2023 1 commit
- Fix lowvram model merging. · a57b0c79
  comfyanonymous authored Aug 26, 2023
  
  a57b0c79
25 Aug, 2023 1 commit
- The new smart memory management makes this unnecessary. · f72780a7
  comfyanonymous authored Aug 25, 2023
  
  f72780a7
24 Aug, 2023 4 commits
- Code cleanups. · 30eb92c3
  comfyanonymous authored Aug 24, 2023
  
  30eb92c3
- Try to free enough vram for control lora inference. · 51dde87e
  comfyanonymous authored Aug 24, 2023
  
  51dde87e
- Always shift text encoder to GPU when the device supports fp16. · cc44ade7
  comfyanonymous authored Aug 23, 2023
  
  cc44ade7
- Even with forced fp16 the cpu device should never use it. · a6ef08a4
  comfyanonymous authored Aug 23, 2023
  
  a6ef08a4
23 Aug, 2023 1 commit
- Save memory by storing text encoder weights in fp16 in most situations. · f081017c
  comfyanonymous authored Aug 23, 2023
```
Do inference in fp32 to make sure quality stays the exact same.
```
  f081017c
20 Aug, 2023 4 commits
- Small cleanups. · 0d7b0a4d
  comfyanonymous authored Aug 20, 2023
  
  0d7b0a4d
- Further tuning and fix mem_free_total. · 92254659
  Simon Lui authored Aug 19, 2023
  
  92254659
- Add ipex optimize and other enhancements for Intel GPUs based on recent memory changes. · 2c096e42
  Simon Lui authored Aug 17, 2023
  
  2c096e42
- --disable-smart-memory now disables loading model directly to vram. · e9469e73
  comfyanonymous authored Aug 20, 2023
  
  e9469e73
17 Aug, 2023 3 commits
- Add --disable-smart-memory for those that want the old behaviour. · 3aee33b5
  comfyanonymous authored Aug 17, 2023
  
  3aee33b5
- Fix issue with regular torch version. · 2be27427
  comfyanonymous authored Aug 17, 2023
  
  2be27427
- Smarter memory management. · 89a0767a
  comfyanonymous authored Aug 17, 2023
```
Try to keep models on the vram when possible.

Better lowvram mode for controlnets.
```
  89a0767a
04 Aug, 2023 1 commit
- Add CMP 30HX card to the nvidia_16_series list. · 1ce0d8ad
  comfyanonymous authored Aug 04, 2023
  
  1ce0d8ad
31 Jul, 2023 1 commit
- Only shift text encoder to vram when CPU cores are under 8. · 4a77fcd6
  comfyanonymous authored Jul 31, 2023
  
  4a77fcd6
30 Jul, 2023 1 commit
- Lower CPU thread check for running the text encoder on the CPU vs GPU. · 3cd31d0e
  comfyanonymous authored Jul 30, 2023
  
  3cd31d0e
23 Jul, 2023 1 commit
- Try to fix memory issue with lora. · 22f29d66
  comfyanonymous authored Jul 22, 2023
  
  22f29d66
19 Jul, 2023 1 commit
- Add MX450 and MX550 to list of cards with broken fp16. · 18885f80
  comfyanonymous authored Jul 19, 2023
  
  18885f80
17 Jul, 2023 2 commits
- Fix device print on old torch version. · ff6b047a
  comfyanonymous authored Jul 17, 2023
  
  ff6b047a
- Add a command line argument to enable backend:cudaMallocAsync · 1679abd8
  comfyanonymous authored Jul 17, 2023
  
  1679abd8
16 Jul, 2023 1 commit
- Lower lora ram usage when in normal vram mode. · 5f573626
  comfyanonymous authored Jul 16, 2023
  
  5f573626
15 Jul, 2023 1 commit
- Speed up lora loading a bit. · 490771b7
  comfyanonymous authored Jul 15, 2023
  
  490771b7
12 Jul, 2023 1 commit
- fix mps miss import · 3e2309f1
  KarryCharon authored Jul 12, 2023
  
  3e2309f1
09 Jul, 2023 1 commit
- Empty cache after model unloading for normal vram and lower. · 0ae81c03
  comfyanonymous authored Jul 09, 2023
  
  0ae81c03
07 Jul, 2023 1 commit
- Add arguments to run the VAE in fp16 or bf16 for testing. · e7bee85d
  comfyanonymous authored Jul 06, 2023
  
  e7bee85d
06 Jul, 2023 1 commit
- Disable autocast in unet for increased speed. · ddc6f12a
  comfyanonymous authored Jul 05, 2023
  
  ddc6f12a
04 Jul, 2023 1 commit
- Fix issue with OSX. · 8d694cc4
  comfyanonymous authored Jul 04, 2023
  
  8d694cc4
03 Jul, 2023 1 commit
- Improvements for OSX. · dc9d1f31
  comfyanonymous authored Jul 03, 2023
  
  dc9d1f31
02 Jul, 2023 2 commits
- Switch to fp16 on some cards when the model is too big. · 2c4e0b49
  comfyanonymous authored Jul 02, 2023
  
  2c4e0b49
- Add a --force-fp16 argument to force fp16 for testing. · 6f3d9f52
  comfyanonymous authored Jul 01, 2023
  
  6f3d9f52
01 Jul, 2023 2 commits
- --gpu-only now keeps the VAE on the device. · 1c1b0e72
  comfyanonymous authored Jul 01, 2023
  
  1c1b0e72
- Leave text_encoder on the CPU when it can handle it. · 3b6fe51c
  comfyanonymous authored Jul 01, 2023
  
  3b6fe51c