Commit Graph
99 Commits
Author SHA1 Message Date
Bubbliiiing 0bf937ea59 Update readme (#281) 2025-08-08 13:32:49 +08:00
Bubbliiiing 5631a8cac4 Update Wan2.2 Fun (#279)
* Update Wan2.2 Fun
2025-08-08 07:54:45 +08:00
Bubbliiiing 1a494b9188 Comyfui Update && Fix bug in api && Dataload (#276)
Update Wan2.2 Comfyui, fix bug in api, fix bug in dataloader, fix bug in readme, fix bug in multigpu
2025-08-05 19:23:02 +08:00
Bubbliiiing 976121891e Merge pull request #274 from aigc-apps/saving
Convert model to weight dtype in saving
2025-08-04 14:49:46 +08:00
Bubbliiiing e7d3b597fb Fix offload bug in wan2.2 inference && Fix boundary bug in training (#271)
* Fix offload bug in wan2.2 inference

* Update boundary
2025-08-01 14:11:23 +08:00
flyingshan ee8c09416b fix(wan2.2 train): Fix high/low noise timestep schedule. (#270) 2025-08-01 13:38:15 +08:00
Bubbliiiing 4dc94d2b95 Update Wan2.2 UI && Add auto_tile_batch_size args in training && Rewrite Wan2.2 init (#269) 2025-07-30 19:54:27 +08:00
NoobNoob 34ef44eb52 docs(Comfyui-README): Fix incorrect ModelScope link for Wan2.1-720P (#268) 2025-07-30 12:59:00 +08:00
Bubbliiiing 9c05f3a741 Fix bug in Wan2.2 (#265) 2025-07-29 20:22:28 +08:00
Bubbliiiing 2d3a5cc7d6 Update Wan2.2 (#264) 2025-07-29 00:06:55 +08:00
Bubbliiiing a17b35acfb Update Torch Compile Node in Comfyui (#253) 2025-07-21 15:22:13 +08:00
Tianhao-Qi 4f14db1269 fix bug in getting mask_pixel_values for non-normal training settings (#252) 2025-07-18 14:13:44 +08:00
Bubbliiiing 9436ff2f6e update discrete_sampler (#249) 2025-07-09 13:49:28 +08:00
Bubbliiiing 9702fc9db6 Merge pull request #246 from aigc-apps/comfyui_json
Update Readme and json
2025-07-04 15:52:53 +08:00
Bubbliiiing 21ec5bf326 Merge pull request #245 from aigc-apps/fix_sage
Set tensor_layout in sageattn
2025-07-04 15:18:51 +08:00
Bubbliiiing dcd1b32ee6 Merge pull request #243 from aigc-apps/controller
Update controller
2025-07-02 18:18:29 +08:00
bubbliiiing 89c8a2a138 Update controller 2025-07-02 10:01:42 +00:00
Bubbliiiing 541f789188 Fix last bug in saving (#241) 2025-07-02 12:14:14 +08:00
huangkunzhe.hkz 9c16b3a73f Set tensor_layout in sageattn 2025-06-27 17:58:41 +08:00
liubo0902and宣源 c2d0c7ae21 add ac (#225)
Co-authored-by: 宣源 <xuanyuan.lb@alibaba-inc.com>
2025-06-26 16:17:03 +08:00
Bubbliiiing 097208e817 Add checkpointing in head && Move gather to the back of the head (#231) 2025-06-25 15:21:53 +08:00
Bubbliiiing 86028a9ce6 Update fix sample size && Fix vram memory bug in wan2.1 lora training (#227) 2025-06-24 10:09:23 +08:00
Bubbliiiing ba1da31fb5 Merge pull request #224 from aigc-apps/support_load_wan_fun_lora
Support loading Wan-Fun LoRAs into Wan base models
2025-06-19 16:17:07 +08:00
Bubbliiiing ce452acc0c Merge pull request #223 from aigc-apps/update_phantom
Update phantom && Fix bug in training readme && Add FSDP Support in Training
2025-06-11 11:52:22 +08:00
bubbliiiing 95e024c681 Update Transformer3d 2025-06-10 05:02:41 +00:00
bubbliiiing a4476549e4 Update Training Code 2025-06-10 03:54:11 +00:00
bubbliiiing 72c2792139 Update Training 2025-06-09 09:42:32 +00:00
bubbliiiing e4c74a12f2 Update phantom && add fsdp training && Fix bug in image encoder while using sage attention 2025-06-09 08:25:49 +00:00
huangkunzhe.hkz fab50bd721 Support loading Wan-Fun LoRAs 2025-06-09 11:17:22 +08:00
Bubbliiiing f176ca9b78 Merge pull request #221 from aigc-apps/fix_merge_lora_multi_gpus
Fix merge_lora in multi gpus
2025-06-06 17:04:49 +08:00
huangkunzhe.hkz 41e0a6a623 Fix merge lora 2025-06-04 20:34:11 +08:00
Bubbliiiing ed5e3e5f69 Merge pull request #220 from aigc-apps/readme_fix_0603
Update Training Readme
2025-06-03 16:00:33 +08:00
bubbliiiing a6a8383c45 Update Training Readme 2025-06-03 07:55:35 +00:00
Bubbliiiing 1061378b66 Add bfloat16 norm && fix bug in cfg_optimization (#217) 2025-05-29 11:28:00 +08:00
Bubbliiiing 6b6413a95c Fix error in comfyui README (#213) 2025-05-26 16:09:37 +08:00
Bubbliiiing a1e6ea6335 Pai sparse attention (#211)
* pai fuser sparse test

* make pai fuser more clear

* make pai fuser more clear

* make pai fuser more clear

* make pai fuser more clear

* Update Readme

* disable compile in rope for less error info

* Fix sage attention backward bug

* Fix checkpoint bugs

* Fix Attention

* Update predict and fast rope

* Fix bug in Sparse

* Fix bug in Sparse

* Fix bug in loras load

* Delete useless import

* Fix bug in ui

* Fix bug in ui
2025-05-26 16:00:58 +08:00
wzgrx 87125a4af0 Update requirements.txt (#208)
Resolve dependency conflicts with other plugins that cause installation failure
2025-05-20 16:33:14 +08:00
Bubbliiiing ed4f10beef Fix bug in lora training && Update zero3 cpu offload && Teacache thresholds (#209) 2025-05-13 20:54:20 +08:00
Bubbliiiing f26f0a809b Update cfg skip to wrapper && Update Teacache && Update Reamde (#200) 2025-05-09 18:11:53 +08:00
Bubbliiiing d7a37ef884 Update dist file download (#193)
* Fix bug in api

* Fix bug in api

* Update dist file download

* Update dist file download
2025-04-30 13:36:09 +08:00
Bubbliiiing cc609b4d87 Fix bug in api (#192)
* Fix bug in api

* Fix bug in api
2025-04-30 10:55:17 +08:00
Bubbliiiing b0fe858bd6 Fix bug in sampler and api. (#191)
* Update sampler and print

* Update sampler and fix some bugs in post
2025-04-30 09:47:01 +08:00
Bubbliiiing d11d6665bc Update sampler and print (#189) 2025-04-29 20:42:21 +08:00
Tianhao-Qi 7d91e1369c fix bugs in i2v training dataset (#185) 2025-04-29 19:08:59 +08:00
Bubbliiiing 7da0e755eb Compile && The model_full_load_and_qfloat8 mode && Move teacache to ui and api. (#188)
* Update Compile

* Update sage

* Update comment

* Update Teacache in ui && Update torch compile

* Update Ref Image in api

* Update ui

* Update ui

* Update ui

* fix bug in ui

* Update ui

* fix bug in ui

* Update longer length in ui

* Fix bug in api check

* Fix bug in pipeline

* Update model_full_load_and_qfloat8

* Update post_to_host

* Update post_to_host
2025-04-29 15:47:30 +08:00
Bubbliiiing 5e2f7fcdac Update Wan Fun V1.1 (#179) 2025-04-27 15:06:36 +08:00
hkz 1fb0696fdc Fix low_cpu_mem_usage with diffusers>=0.33.0 (#181) 2025-04-27 14:57:55 +08:00
Tianhao-Qi faa208e8d9 Add FlowDPMSolverMultistepScheduler, FlowUniPCMultistepScheduler, keep consistent with original teacache and avoid unnecessary computation (#175) 2025-04-27 14:08:13 +08:00
mohui37 3c19c391cd Update utils.py (#169) 2025-04-25 20:25:29 +08:00
Tianhao-Qi 278a4c3cd2 fix bugs in if_else conditions for function get_teacache_coefficients (#173) 2025-04-17 19:06:37 +08:00