nodes/pollockjj-ComfyUI-MultiGPU
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
125 Commits 4 Branches 0 Tags
2d81ef0a219161090ac0974bec70a463f4519436
Commit Graph
125 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
John Pollock 2d81ef0a21 Support for kijai's ComfyUI-WanVideoWrapper 2025-03-23 13:40:05 -05:00
John Pollock 1bf9333fc7 chore: update version to 1.6.2 and fix description formatting in pyproject.toml 2025-02-12 12:18:13 -06:00
John Pollock a2093a4fc9 feat: add text encoder device handling, whereas CLIP can sometimes default to CPU, whereas using a DisTorch CLIP load you can load the layes on CPU buy use CUDA for processing. Especially helpful llava-llama 2025-02-12 11:51:58 -06:00
John Pollock 6500ca9a47 docs: enhance README for clarity on DisTorch Virtual VRAM features and usage based on user success stories and improved ease of use 2025-02-11 22:50:03 -06:00
John Pollock c431a7d870 Merge pull request #16 from eltociear/patch-1 
docs: update README.md
 2025-02-09 19:21:30 -06:00
Ikko Eltociear Ashimine e8be359b35 docs: update README.md 
promot -> prompt
 2025-02-10 03:25:20 +09:00
John Pollock 62646d3ca3 Push Distorch 2.0 Virtual VRAM release out to Comfy Registry 2025-02-07 22:00:48 -06:00
John Pollock 04882505cb Merge dev into main, taking dev version of __init__.py 2025-02-07 21:52:58 -06:00
John Pollock c01ef265a8 Update README to enhance clarity on manual allocation strings and installation instructions 2025-02-07 21:46:01 -06:00
John Pollock e36cec9fb7 Add new assets and update README for DisTorch 2.0 features 2025-02-07 21:45:41 -06:00
John Pollock 375276a485 Add files via upload 
preparing for DisTorch 2.0 launch
 2025-02-07 21:26:07 -06:00
John Pollock 1af932be16 Add files via upload 
preparing for DisTorch 2.0 launch
 2025-02-07 20:34:50 -06:00
John Pollock 8642f75a12 Add files via upload 2025-02-07 20:17:37 -06:00
John Pollock c8c4e74692 cleaning up experimental workflows 2025-02-07 19:17:06 -06:00
John Pollock 9bd984b420 Update default value for virtual VRAM GB to 4.0 in override_class_with_distorch 2025-02-07 18:27:14 -06:00
John Pollock de2219c974 Refactor virtual VRAM allocation logic and improve logging format 2025-02-07 18:24:28 -06:00
John Pollock c98a535435 Refactor logging in DisTorch analysis and update allocation handling for virtual VRAM 2025-02-07 16:09:35 -06:00
John Pollock 3a4c6d50c8 Virtual VRAM "automatic" mode for DisTorch, WIP but working 2025-02-07 15:05:08 -06:00
John Pollock 5a403e638c MergeFluxLoRAsQuantizeAndLoad, WIP 2025-02-07 04:43:45 -06:00
John Pollock 1ffa88f271 Benchmarking JSON for Reddit user 2025-02-06 14:29:26 -06:00
John Pollock 08e3bbfcb0 add example for reddit user 2025-02-06 11:53:05 -06:00
John Pollock d0d33a69ac Corrected Florence2 nodes and removed DiffSynth to remain consistent with emergency release to main 2025-02-06 03:09:00 -06:00
John Pollock 3679c268ac 🔧 BREAKING CHANGE: DiffSynth functionality has been temporarily removed while investigating device management issue (Having two GPUs with different amounts of VRAM leads to OOMs pollockjj/ComfyUI-MultiGPU#13). 
If you were using nodes with DiffSynth in their name (like ...DiffSynthMultiGPU), please switch to the standard MultiGPU versions for now (e.g., ...MultiGPU). This change eliminates a device management issue that was affecting some Windows users.

See: https://github.com/pollockjj/ComfyUI-MultiGPU/issues/13

Most users won't be affected as this only impacts the DiffSynth variants of nodes.

If you need help modifying your workflows, please open an issue. Will revisit DiffSynth functionality once issue can be contained or worked-around

Update comfyregistry to 1.5.0 to reflect major change in functionality, in this case, a reduction.
 2025-02-06 00:32:21 -06:00
John Pollock 4182275f85 Multiple LoRA DisTorch example 2025-02-05 10:55:17 -06:00
John Pollock 4a8d70a0d4 refactored to move stable wrapper nodes into nodes.py and remainder in init.py 2025-02-03 09:15:05 -06:00
John Pollock 262ceea716 Add precompiled llama-quantize binaries for flux Just-in-time GGUF quantization 
These binaries are built from https://github.com/pollockjj/llama.cpp/tree/flux-quant-b3600

Linux binary (SHA256: 846c0bab3c7f7c6729f22b6229a2db5da2da63a4549ef8bf76546092f56fec15):
- Ubuntu 22.04
- gcc 13.3.0
- Debug build with flags: --config Debug -j10 --target llama-quantize

Windows binary (SHA256: 121c02184fcf30dc4ff3dcf4a69177e44aced9d7ba8c77f49cd138c2d73f7e6b):
- Windows 11
- MSVC 19.42.34436.0
- Debug build with flags: -G "Visual Studio 17 2022" -A x64 -DBUILD_SHARED_LIBS=OFF

Binaries are verified and released at:
https://github.com/pollockjj/llama.cpp/releases/tag/1.0.0
 2025-02-03 00:15:18 -06:00
John Pollock df779bf820 Add new example JSON for Maleficent_End8551 workflow 2025-02-01 12:35:49 -06:00
John Pollock f5038d73bd bitsandbytes_NF4 example workflows 2025-02-01 08:00:00 -06:00
John Pollock 11d41b5192 Example of taking lowvram flow even lower 2025-02-01 06:19:40 -06:00
John Pollock 89ded6cc48 Added new ltxvideo example for lowvram, high dram - run ~5 minutes for a 5 second video on my 3090 2025-01-31 11:42:49 -06:00
John Pollock 3e130e3dfb Remove log_comfy_states function - no longer needed 2025-01-31 06:45:43 -06:00
John Pollock 005b5b1882 This release includes an embeddings adapter for the IP2V part of kijai's CLIP loader for HunyuanVideo. See examples. Bump version to 1.4.3 and update category for HunyuanVideoEmbeddingsAdapter to multigpu; enhance README with new workflow examples for HunyuanVideo GGUF-quantized models. 2025-01-29 11:55:16 -06:00
John Pollock 3260b7e38e Add HunyuanVideoEmbeddingsAdapter class for using kijai's IP2V conditioning video embeddings in the standard sampler, allowing it to be used with GGUF/DisTorch methods. 2025-01-29 09:25:19 -06:00
John Pollock 185ecc2ab5 Push fix to check_module_exists() to Comfy Registry, thanks @3dluvr 2025-01-29 05:16:45 -06:00
John Pollock fc4e2b2f12 Merge pull request #12 from 3dluvr/main 
Fix check_module_exists() to use folder_paths
 2025-01-29 05:12:49 -06:00
3dluvr 379ecce687 Fix check_module_exists() to use folder_paths 
In Windows, module detection was failing because the method couldn't find the hard-coded custom_nodes/ folder in os.join.path.

We switch to using folder_paths which will return a correct path regardless of the platform.
 2025-01-28 19:59:44 -05:00
John Pollock 41e974f086 1.40 -->1.4.1 
Pop 1.4 (DisTorch) to Comfy Registry
 2025-01-28 15:41:13 -06:00
John Pollock 369255ff8d Merge branch 'main' of https://github.com/pollockjj/ComfyUI-MultiGPU into dev 2025-01-28 06:26:25 -06:00
John Pollock 5ff6a90b17 preparing DisTorch for release to :main: 2025-01-28 06:06:53 -06:00
John Pollock 8f3ff06fa3 rename files for consistency, new DisTorch examples 2025-01-28 05:19:50 -06:00
pollock c07a345c45 Fix case sensitivity in module check for HunyuanVideoWrapper in __init__.py 2025-01-27 18:32:20 -05:00
John Pollock 7ccea97c52 Update device allocation format and enhance module check for case insensitivity in __init__.py 2025-01-27 17:24:15 -06:00
John Pollock c74f66b9ad icon 2025-01-27 23:06:56 -06:00
John Pollock 3dbfcc7135 Chasing down bug causing incorrect patched device with distorch code. Re-integrated distorch into __init__.py as one of the consequences. 2025-01-25 17:49:12 -06:00
John Pollock d67b18474a Merge branch 'dev' of https://github.com/pollockjj/ComfyUI-MultiGPU into dev 2025-01-24 21:06:10 -06:00
John Pollock b99e152686 Refactor GGUF model patcher to enhance device handling and introduce model hashing for allocation tracking 2025-01-24 21:06:01 -06:00
pollock 101fb49848 Refactor model hash creation and logging for clarity; remove unnecessary logging statements in distorch.py 2025-01-24 11:00:38 -05:00
pollock ef614f840e Move distorch parameter storage from model attachment (non-working) to global table with hash. 2025-01-24 09:22:43 -05:00
pollock 2ac057d338 Implement model hash creation and allocation storage for MultiGPU support in GGUF model patcher 2025-01-24 05:37:17 -05:00
John Pollock 624c893942 Refactored into init.py and distorch.py 2025-01-23 13:16:18 -06:00