Jaret Burkett
065ac27353
Fix issue with compiling convrot on some models
2026-07-11 09:26:12 -06:00
Jaret Burkett
96a3a06111
Added initial experimental support for convrot4 and convrot8 quantization.
2026-07-10 11:01:14 -06:00
Jaret Burkett
6fac83d068
Fix casting issue with orbit quant
2026-07-10 10:23:38 -06:00
Jaret Burkett
71c75357eb
Add cached conditioning recovery to wan 22 5b model
2026-07-10 10:22:24 -06:00
Jaret Burkett
ad07b06de5
Use cached first frame for wan22_5 model
2026-07-10 10:21:34 -06:00
Jaret Burkett
886c2aec57
Allow for vae tiling onle without low vram on wan models with a model kwarg
2026-07-10 09:42:52 -06:00
Jaret Burkett
fe82487187
Add tiling on vae decode for qwen image models when low vram flag is on
2026-07-10 09:07:29 -06:00
Jaret Burkett
e7951ad29e
Exclude sensative layers from quantization on wan models
2026-07-10 08:56:05 -06:00
Jaret Burkett
883d60eb71
Do vae tiling when decoding wan models with low vram active.
2026-07-10 08:02:32 -06:00
Jaret Burkett
fed9357234
Fix ui to show kv cache for krea2 edit raw
2026-07-09 15:48:46 -06:00
Jaret Burkett
5a9b5bde3f
Added experimental orbit quant
2026-07-09 15:14:51 -06:00
Jaret Burkett
a4bbe167ce
Added reference token attention isolation (kv_cache) for Krea2 edit training. Same training cost with significant inference speed up. 2x inference speedup.
2026-07-09 12:00:55 -06:00
Jaret Burkett
6233efe1bb
Fix sampling issue on zimage turbo and krea2 turbo models
2026-07-08 12:34:39 -06:00
Jaret Burkett
dd08579eda
Add ability to pull control images from same folder group
2026-07-08 05:21:31 -06:00
Jaret Burkett
7bceec3b07
Apply a loss mask for first frame conditioning for wan and ltx
2026-07-07 09:05:53 -06:00
Jaret Burkett
bd93a312bc
Load more types of comfy ui style quants for ltx2
2026-07-07 09:04:51 -06:00
Jaret Burkett
17bc302d13
Make log calls blocking to prevent duplicates.
2026-07-06 07:38:20 -06:00
Jaret Burkett
6c0d1c4679
Improve latency of the job list endpoint
2026-07-04 13:22:09 -06:00
Jaret Burkett
3a94591c89
Disable text encoder unload with krea2 o-edit
2026-07-04 09:17:15 -06:00
Jaret Burkett
b1e1a834d4
Added support to train Krea2 as an edit model
2026-07-04 09:12:39 -06:00
Jaret Burkett
f63221e577
Replace all sync functions with async to allow more parallel api calls
2026-07-01 17:12:43 -06:00
Jaret Burkett
48781f900b
Improve the loading and transfer speed of the dataset file lists
2026-07-01 16:40:01 -06:00
Jaret Burkett
b36a8e9c4b
Only pull the new log bytes from the ui to improve performance
2026-07-01 14:11:23 -06:00
Jaret Burkett
733e14cb58
Fix issue where the window would scroll to and hilight custom items in the select box
2026-07-01 13:57:56 -06:00
Jaret Burkett
4e50535478
Allow models to provide an additional loss
2026-06-28 09:49:36 -06:00
Jaret Burkett
1e12b6b73f
Remove duplicate save when triggering save next step
2026-06-28 07:18:16 -06:00
Jaret Burkett
7ee1f98f6d
Allow loading and saving z_image in comfy safetensors format.
2026-06-28 07:15:49 -06:00
Jaret Burkett
c97fc9973a
Fix loading pretraied lora when merging network on save
2026-06-28 06:04:36 -06:00
Jaret Burkett
f8667f0334
Save Krea2 even if quantized
2026-06-28 06:03:37 -06:00
Jaret Burkett
df6ea4263d
Version bump
2026-06-26 13:02:10 -06:00
Jaret Burkett
ad87aacec0
Add ability to set certain layers to full for loras
2026-06-26 13:01:48 -06:00
Jaret Burkett
4a99ddabad
Fix breaking change with diffusers qwen image
2026-06-26 10:04:35 -06:00
Jaret Burkett
5f04ae7ad5
Rework merge_network_on_save to handle dequantization on merging and saving for much more efficient full finetuning.
2026-06-25 13:19:38 -06:00
Jaret Burkett
6ecff36f26
Add a way to do full rank lora modules of non lora layers such as embeddings, norm, etc
2026-06-25 11:55:07 -06:00
Jaret Burkett
4eb0707639
Add a control generation script.
2026-06-25 10:14:24 -06:00
Jaret Burkett
d14f6e567a
Allow individual models to scale the loss after it is calculated.
2026-06-25 10:13:58 -06:00
Jaret Burkett
f743ccf7ef
Save before last sample
2026-06-24 09:13:01 -06:00
Jaret Burkett
7a089fd0d7
Add support for training directly on Krea2 Turbo with a training adapter
2026-06-23 20:32:02 -06:00
Jaret Burkett
a803611ec1
Enable tiling on vae when decoding with low_vram flag on krea2
2026-06-23 18:38:52 -06:00
Jaret Burkett
724e67d634
Set krea 2 to use new lokr format
2026-06-23 13:31:50 -06:00
Jaret Burkett
e20b42e84a
Add offloading support for krea2
2026-06-23 10:43:36 -06:00
Jaret Burkett (Ostris)
99be3d96a2
Add support for Krea2 ( #906 )
...
* Add support for krea2
* Update repo pointer to actual repo
2026-06-23 09:18:14 -06:00
Jaret Burkett
af594061ab
Add ability to do hidden states with tipsv2
2026-06-22 11:00:04 -06:00
Jaret Burkett
820d534d6e
Add features for models that may need a non masked loss such as inpainiting.
2026-06-22 10:59:17 -06:00
Jaret Burkett
c133c55cf5
Fix issue with mask generator revision
2026-06-22 08:19:27 -06:00
Jaret Burkett
d51463ca52
Add ability to recover from a truncated image file in the dataset
2026-06-22 05:07:30 -06:00
Jaret Burkett
ba0b3dbb65
Force batch size when bucket is too small by duplicating items in the batch
2026-06-21 20:01:29 -06:00
Jaret Burkett
dba092fc15
Keep a real git repo in the docker image
2026-06-21 06:54:19 -06:00
Jaret Burkett
548a286992
Add captioning monitoring information to the dataset page so you can just stay on that page with all the info there.
2026-06-19 08:00:40 -06:00
Jaret Burkett
99f8fd44e3
Add a fallback bbox adjustment when json parsing fails on ideogram4 captioner
2026-06-19 07:25:54 -06:00
Jaret Burkett
022d1c29e0
Remove color pallet for individual objects in the caption prompts
2026-06-19 06:56:18 -06:00
Jaret Burkett
60c1ac6a50
Add support for Boogu Image and Boogu Image Edit
2026-06-18 15:05:49 -06:00
Jaret Burkett
e886745051
Added additional information on addine new models and some additional gotchas
2026-06-18 15:04:50 -06:00
Jaret Burkett
515b0ea5cd
Remove transformer log supression from captioner
2026-06-18 10:00:30 -06:00
Jaret Burkett
e8c828089a
Fix issue where gpu sometimes doesnt show on caption modal
2026-06-18 09:37:59 -06:00
Jaret Burkett
ad49d4ef25
Update example to cover some common issues
2026-06-18 07:54:37 -06:00
Jaret Burkett
66f7c06742
Patch away the qwen3vl conv3d that only has slow bf16 kernels
2026-06-17 14:29:20 -06:00
Jaret Burkett
92814f9e6d
Apply ideogram dynamic shifting to sampling
2026-06-16 18:52:16 -06:00
Jaret Burkett
178eb5fbbe
Add unconditional lora support so Ideogram 4 inference will more closely resemble the full pipeline results. I pushed a finetuned unconditional lora to the hub as an adapter.
2026-06-16 13:27:43 -06:00
Jaret Burkett
f6c0104f25
Handle unconditional conditioning for ideogram 4 more in line with example code.
2026-06-16 10:23:08 -06:00
Jaret Burkett
86b19589a0
Update the Ideogram 4 prompt generation/parsing/ui to handle the updated format notes better.
2026-06-16 09:44:38 -06:00
Jaret Burkett
fcccc0fbd2
Add gradient checkpointing to ideogram4 vae
2026-06-15 10:00:40 -06:00
Jaret Burkett
c730d64478
Added a flag to keep loading the image when latents are cached. Useful for DFE and other methods that target pixelspace losses.
2026-06-15 05:31:48 -06:00
Jaret Burkett
faa770fc79
Another docker thing?
2026-06-14 12:27:05 -06:00
Jaret Burkett
570c806924
Mode docker build work
2026-06-14 12:24:55 -06:00
Jaret Burkett
ebbb09230b
Add requirements base to docker build
2026-06-14 12:21:28 -06:00
Jaret Burkett
5df3fb69e3
Rework Docker image for a minimal build/pull/push footprint
2026-06-14 12:19:58 -06:00
Jaret Burkett
c0d600b5d6
Allow using flash backend for ideogram
2026-06-13 15:17:20 -06:00
Jaret Burkett
c8cd78b1a4
Allow nested transformer block names for quantization, lora targeting, quantizing
2026-06-13 14:33:16 -06:00
Jaret Burkett
17c9279828
Version bump
2026-06-13 09:48:26 -06:00
Jaret Burkett
6c3b82696e
Add support for PRX Pixel T2I
2026-06-13 09:47:53 -06:00
Jaret Burkett
a01c83073a
Added an example model with docs so people and agents can add models easier.
2026-06-13 08:17:06 -06:00
Jaret Burkett
2f91db8363
Defauly to compiling full graph to false
2026-06-13 07:28:51 -06:00
Jaret Burkett
0165fb2ac6
Update npm packages
2026-06-12 17:50:47 -06:00
Jaret Burkett
c90c400716
Add Ustris Cloud info to the README
2026-06-12 12:53:47 -06:00
Jaret Burkett
43b22b91ee
Add a triton check on compile
2026-06-12 12:26:16 -06:00
Jaret Burkett
10e50d5797
Fix issue with casting unet after compilation
2026-06-12 12:16:51 -06:00
Jaret Burkett
d83f7dd4d9
Fix a few issues with compile. Changed defaults. Future proofed block layer compile.
2026-06-12 11:43:43 -06:00
Jaret Burkett
a5558ae7d9
Add compile to the actual right config section.
2026-06-12 11:12:41 -06:00
Jaret Burkett
c09b228a35
Add model compiling to the ui
2026-06-12 11:02:10 -06:00
Jaret Burkett
0f580f0663
Add compiling option for captioning. Added progress bar for captioning on active job widget. Track steps on caption. Added reverse proxy iframe that plugins can use to add ui functionality.
2026-06-12 10:26:52 -06:00
Jaret Burkett
3fd14f3805
Change suporters view to an auto updating SVG file so the README doesnt need to be updated constantly.
2026-06-12 09:48:50 -06:00
Jaret Burkett
9cf34f945c
Allow multi selecting jobs and deleting them in one go
2026-06-12 08:01:56 -06:00
Jaret Burkett
55ce6570f2
Automagic 3 rework. Stable in my testing.
2026-06-12 07:52:44 -06:00
Jaret Burkett
88127557f5
Comment out logging supression that was keeping weight loadings from being shown with transformers library
2026-06-11 07:53:41 -06:00
Jaret Burkett
01b6a9806b
Another complete rework of automagic3. Added a decay to the LR spread to the mean to prevent LRs fighting with eachother
2026-06-09 12:08:45 -06:00
Jaret Burkett
9e99d3ce5d
Save loss graph view settings per page to local storage.
2026-06-09 07:29:51 -06:00
Jaret Burkett
acb1548722
Updated the comments and doc for Automagic v3
2026-06-09 07:05:48 -06:00
Jaret Burkett
a1ac6e8b01
Reworked automagic v3 again. Seems more stable. Still testing.
2026-06-08 22:03:54 -06:00
Jaret Burkett
5d6887fd98
Major updates to automagic3 optimizer. Seems to be functioning more ideally and naturally decays, as it should.
2026-06-08 14:09:02 -06:00
Jaret Burkett
0d018db689
Rework the toggles so we can hide the trend line from the loss graph. No need for a raw toggle either.
2026-06-08 13:56:37 -06:00
Jaret Burkett
cac3815b2c
Make DOP run in a single backward pass, should be faster and more stable. Show dop loss on ui
2026-06-08 13:55:47 -06:00
Jaret Burkett
687def6f7a
Add better smoothing with ema rounding of the ends of the loss graph for first and last instance do not dominate the smoothing factors.
2026-06-08 10:43:46 -06:00
Jaret Burkett
d7f8887bbf
Show all logged metrics on the loss graph
2026-06-08 09:43:03 -06:00
Jaret Burkett
e281df70dd
Allow automagic3 to run unfused. Add some clipping.
2026-06-08 08:59:06 -06:00
Jaret Burkett
c9cdbb5bb7
Version bump
2026-06-07 16:08:11 -06:00
Jaret Burkett
c78b1404e3
Deepen offload prefetch pipeline with per-slot events
...
Replace the 2-slot ping-pong + single global "compute-started" event
with a depth-N ring buffer where each transfer waits only on the slot
it's reusing (D layers back) instead of the most-recent compute. Applies
to forward and backward, Linear and Conv. Depth is tunable via
AI_TOOLKIT_OFFLOAD_DEPTH (default 4).
Bit-exact vs non-offload (output, grad_input, weight grads). No speedup
on a bandwidth-bound PCIe link (already saturated at depth 2), but the
cleaner per-slot design removes the fragile shared-event serialization
and lets deeper prefetch help on faster buses.
2026-06-07 16:07:13 -06:00
Jaret Burkett
cdff6e36aa
Pin inner stores of torachao to speed up layer offloading for quantized models around 25%
2026-06-07 15:52:01 -06:00
Jaret Burkett
75781fb5a5
Fix float8 weights not offloading to CPU in layer offloading
2026-06-07 15:36:17 -06:00
Jaret Burkett
35588726de
Fixed issue where a buffer was stuck on cpu when offloading ideogram4
2026-06-07 13:31:03 -06:00
Jaret Burkett