araleza
|
cd239f0fa9
|
Moved kahan state from file globals to optimizer state variables
|
2025-08-20 16:42:15 +01:00 |
|
araleza
|
648994271e
|
Added log output message to show that Kahan summation is being used
|
2025-07-29 10:28:26 +01:00 |
|
araleza
|
3f0230a286
|
Now sending int16s instead of f32s to cpu device; faster and maybe more accurate
|
2025-07-29 10:05:06 +01:00 |
|
araleza
|
acb4cf32e8
|
Fixed a warning typo, and changed --kahan-summation to --kahan_summation
|
2025-07-23 18:25:07 +01:00 |
|
araleza
|
bb7750fbca
|
Fixed typo in comment
|
2025-07-23 15:10:57 +01:00 |
|
araleza
|
da6416a2fc
|
Restoring the deleted __main__ function and fixing a warning typo
|
2025-07-23 15:08:24 +01:00 |
|
araleza
|
6517b2b838
|
Added support for Kahan summation for Adafactor-optimized Flux FFT
|
2025-07-23 14:34:32 +01:00 |
|
Kohya S.
|
4987057701
|
Merge pull request #2042 from kohya-ss/update-docs
experimental: Documentation Updates with Gemini
|
2025-07-21 13:48:34 +09:00 |
|
Kohya S
|
c84a163b32
|
docs: update README for documentation
|
2025-07-21 13:40:03 +09:00 |
|
Kohya S
|
7de68c1eb1
|
Merge branch 'sd3' into update-docs
|
2025-07-21 13:32:43 +09:00 |
|
Kohya S.
|
d98400b06e
|
Merge pull request #2138 from kohya-ss/feature-lumina-image
Feature lumina image
|
2025-07-21 13:21:26 +09:00 |
|
Kohya S
|
518545bffb
|
docs: add support information for Lumina-Image 2.0 in recent updates
|
2025-07-21 13:16:42 +09:00 |
|
Kohya S
|
d300f19045
|
docs: update Lumina training guide to include inference script and options
|
2025-07-21 13:15:09 +09:00 |
|
Kohya S
|
aec7e16094
|
feat: add an option to add system prompt for negative in lumina inference
|
2025-07-21 13:14:59 +09:00 |
|
Kohya S.
|
d53a532a82
|
Merge pull request #2153 from rockerBOO/fix-alphas-cumprod
Fix alphas cumprod after add_noise for DDIMScheduler
|
2025-07-17 21:37:55 +09:00 |
|
Dave Lage
|
3adbbb6e33
|
Add note about why we are moving it
|
2025-07-16 16:09:20 -04:00 |
|
rockerBOO
|
a7b33f3204
|
Fix alphas cumprod after add_noise for DDIMScheduler
|
2025-07-15 22:36:46 -04:00 |
|
Kohya S
|
c0c36a4e2f
|
fix: remove duplicated latent normalization in decoding
|
2025-07-15 21:58:03 +09:00 |
|
Kohya S
|
25771a5180
|
fix: update help text for cfg_trunc_ratio argument
|
2025-07-15 21:53:13 +09:00 |
|
Kohya S
|
e0fcb5152a
|
feat: support Neta Lumina all-in-one weights
|
2025-07-15 21:34:35 +09:00 |
|
Kohya S
|
13ccfc39f8
|
fix: update flow matching loss and variable names
|
2025-07-13 21:26:06 +09:00 |
|
Kohya S
|
30295c9668
|
fix: update parameter names for CFG truncate and Renorm CFG in documentation and code
|
2025-07-13 21:00:27 +09:00 |
|
Kohya S
|
999df5ec15
|
fix: update default values for timestep_sampling and model_prediction_type in training arguments
|
2025-07-13 20:52:00 +09:00 |
|
Kohya S
|
88960e6309
|
doc: update lumina LoRA training guide
|
2025-07-13 20:49:38 +09:00 |
|
Kohya S
|
88dc3213a9
|
fix: support LoRA w/o TE for create_network_from_weights
|
2025-07-13 20:46:24 +09:00 |
|
Kohya S
|
1a9bf2ab56
|
feat: add interactive mode for generating multiple images
|
2025-07-13 20:45:09 +09:00 |
|
Kohya S
|
8a72f56c9f
|
fix: clarify Flash Attention usage in lumina training guide
|
2025-07-11 22:14:16 +09:00 |
|
kohya-ss
|
d0b335d8cf
|
feat: add LoRA training guide for Lumina Image 2.0 (WIP)
|
2025-07-10 20:15:45 +09:00 |
|
kohya-ss
|
2a53524588
|
Merge branch 'sd3' into update-docs
|
2025-07-10 19:40:33 +09:00 |
|
kohya-ss
|
4e7dfc0b1b
|
Merge branch 'sd3' into feature-lumina-image
|
2025-07-10 19:40:12 +09:00 |
|
Kohya S.
|
2e0fcc50cb
|
Merge pull request #2148 from kohya-ss/update-gitignore-for-claude-and-gemini
feat: add .claude and .gemini to .gitignore
|
2025-07-10 19:37:43 +09:00 |
|
kohya-ss
|
0b90555916
|
feat: add .claude and .gemini to .gitignore
|
2025-07-10 19:34:31 +09:00 |
|
Kohya S.
|
9a50c96a68
|
Merge pull request #2147 from kohya-ss/ai-coding-agent-prompt
Add prompt guidance files for Claude and Gemini
|
2025-07-10 19:25:11 +09:00 |
|
kohya-ss
|
7bd9a6b19e
|
Add prompt guidance files for Claude and Gemini, and update README for AI coding agents
|
2025-07-10 19:16:05 +09:00 |
|
Kohya S
|
3f9eab4946
|
fix: update default values in lumina minimal inference as same as sample generation
|
2025-07-09 23:33:50 +09:00 |
|
Kohya S
|
7fb0d30feb
|
feat: add LoRA support for lumina minimal inference
|
2025-07-09 23:28:55 +09:00 |
|
Kohya S
|
b4d1152293
|
fix: sample generation with system prompt, without TE output caching
|
2025-07-09 21:55:36 +09:00 |
|
Kohya S.
|
2fffcb605c
|
Merge pull request #2146 from rockerBOO/lumina-typo
Lumina: Change to 3
|
2025-07-08 21:49:11 +09:00 |
|
rockerBOO
|
a87e999786
|
Change to 3
|
2025-07-07 17:12:07 -04:00 |
|
Kohya S
|
05f392fa27
|
feat: add minimum inference code for Lumina with image generation capabilities
|
2025-07-03 21:47:15 +09:00 |
|
Kohya S
|
6731d8a57f
|
fix: update system prompt handling
|
2025-06-29 22:21:48 +09:00 |
|
Kohya S
|
078ee28a94
|
feat: add more workaround for 'gated repo' error on github actions
|
2025-06-29 22:06:19 +09:00 |
|
Kohya S
|
5034c6f813
|
feat: add workaround for 'gated repo' error on github actions
|
2025-06-29 22:00:58 +09:00 |
|
Kohya S
|
884c1f37c4
|
fix: update to work with cache text encoder outputs (without disk)
|
2025-06-29 21:58:43 +09:00 |
|
Kohya S
|
935e0037dc
|
feat: update lumina system prompt handling
|
2025-06-29 21:33:09 +09:00 |
|
Kohya S.
|
52d13373c0
|
Merge pull request #1927 from sdbds/lumina
Support Lumina-image-2.0
|
2025-06-29 20:01:24 +09:00 |
|
Qing Long
|
8e4dc1f441
|
Merge pull request #28 from rockerBOO/lumina-train_util
Lumina train util
|
2025-06-17 16:06:48 +08:00 |
|
rockerBOO
|
0e929f97b9
|
Revert system_prompt for dataset config
|
2025-06-16 16:50:18 -04:00 |
|
rockerBOO
|
1db78559a6
|
Merge branch 'sd3' into update-sd3
|
2025-06-16 16:43:34 -04:00 |
|
Kohya S.
|
3e6935a07e
|
Merge pull request #2115 from kohya-ss/fix-flux-sampling-accelerate-error
Fix unwrap_model handling for None text_encoders in sample_images
|
2025-06-15 21:14:09 +09:00 |
|