Ryan Dick
477d87ec31
Fix layer patch dtype selection for CLIP text encoder models.
2024-12-29 21:48:51 +00:00
Ryan Dick
6d7314ac0a
Consolidate the LayerPatching patching modes into a single implementation.
2024-12-24 15:57:54 +00:00
Ryan Dick
80db9537ff
Rename model_patcher.py -> layer_patcher.py.
2024-12-24 15:57:54 +00:00
Ryan Dick
61253b91f1
Enable LoRAPatcher.apply_smart_lora_patches(...) throughout the stack.
2024-12-24 15:57:54 +00:00
Ryan Dick
7dc3e0fdbe
Get rid of ModelLocker. It was an unnecessary layer of indirection.
2024-12-24 14:23:18 +00:00
Ryan Dick
d30a9ced38
Rename model_cache_default.py -> model_cache.py.
2024-12-24 14:23:18 +00:00
Ryan Dick
e0bfa6157b
Remove ModelCacheBase.
2024-12-24 14:23:18 +00:00
Ryan Dick
83ea6420e2
Move CacheStats to its own file.
2024-12-24 14:23:18 +00:00
Riku
525cb38c71
fix(app): fixed InputField default values
2024-12-20 09:30:56 +11:00
psychedelicious
a9a6720bad
feat(app): change queue item execution log from debug to info
...
This provides useful context for subsequent logs during queue item execution.
2024-12-20 09:19:04 +11:00
psychedelicious
858bf9cf8c
feat(api): less verbose uvicorn logs
...
Uvicorn's logging is rather verbose. This change adds a `log_level_network` config setting to independently control uvicorn's log outputs. The setting defaults to warning.
The change hides the helpful startup message that says the host and port we are running on.
For example: `Uvicorn running on http://0.0.0.0:9090 (Press CTRL+C to quit`
The ASGI lifespan handler is updated to log an equivalent message on startup, regardless of log level settings.
Besides being helpful, the launcher relies on a message like this to launch the app. So, previously, if the user set their log level to anything above info (e.g. warning or error), the launcher would fail to open the app. This change prevents that edge case.
2024-12-20 09:19:04 +11:00
David Hauptman
74a29c3735
re-format to fix ruff error
2024-12-19 22:33:17 +11:00
David Hauptman
6fc6be3aa0
Fix error message when adding a local path with quotes around the string
2024-12-19 22:33:17 +11:00
Mary Hipp
c154d833b9
raise error if control lora used with schnell
2024-12-18 10:19:28 -05:00
Ryan Dick
a463e97269
Bump FluxControlLoRALoaderInvocation version.
2024-12-17 13:36:10 +00:00
Ryan Dick
b272d46056
Enable ability to control the weight of FLUX Control LoRAs.
2024-12-17 13:36:10 +00:00
Ryan Dick
dd09509dbd
Rename ModelPatcher -> LayerPatcher to avoid conflicts with another ModelPatcher definition.
2024-12-17 13:20:19 +00:00
Ryan Dick
7fad4c9491
Rename LoRAModelRaw to ModelPatchRaw.
2024-12-17 13:20:19 +00:00
Ryan Dick
b820862eab
Rename ModelPatcher methods to reflect that they are general model patching methods and are not LoRA-specific.
2024-12-17 13:20:19 +00:00
Ryan Dick
c604a0956e
Rename LoRAPatcher -> ModelPatcher.
2024-12-17 13:20:19 +00:00
Ryan Dick
41664f88db
Rename backend/patches/conversions/ to backend/patches/lora_conversions/
2024-12-17 13:20:19 +00:00
Ryan Dick
42f8d6aa11
Rename backend/lora/ to backend/patches
2024-12-17 13:20:19 +00:00
Ryan Dick
a4bed7aee3
Minor tidy of FLUX control LoRA implementation. (mostly documentation)
2024-12-17 07:28:45 -05:00
Ryan Dick
d84adfd39f
Clean up FLUX control LoRA pre-processing logic.
2024-12-17 07:28:45 -05:00
Ryan Dick
ac82f73dbe
Make FluxControlLoRALoaderOutput.control_lora non-optional.
2024-12-17 07:28:45 -05:00
Brandon Rising
046d19446c
Rename Structural Lora to Control Lora
2024-12-17 07:28:45 -05:00
Brandon Rising
f53da60b84
Lots of updates centered around using the lora patcher rather than changing the modules in the transformer model
2024-12-17 07:28:45 -05:00
Brandon Rising
5a035dd19f
Support bnb quantized nf4 flux models, Use controlnet vae, only support 1 structural lora per transformer. various other refractors and bugfixes
2024-12-17 07:28:45 -05:00
Brandon Rising
f3b253987f
Initial setup for flux tools control loras
2024-12-17 07:28:45 -05:00
Eugene Brodsky
4c94d41fa9
(chore) ruff format
2024-12-04 17:02:08 +00:00
Eugene Brodsky
4036244ee9
(app) clarify log message when migrating old .cache
2024-12-04 17:02:08 +00:00
Eugene Brodsky
d06232d9ba
(config) ensure legacy model configs and node template are writable by the user even if the source files are read-only
2024-12-04 17:02:08 +00:00
psychedelicious
3305bad0c2
fix(app): queue item id check before setting cancel flag should use != instead of is not
...
The `is` operator compares references, not values. Thanks to a wonderfully unintuitive quirk of python, `is` works on integers from `-5` to `256`, inclusive.
Whenever integers in this range are used for a value, internally python returns a reference to a stable object in memory. When integers outside this range are used as a value, python creates a new object in memory for that integer.
See `PyLong_FromLong` documentation here: https://docs.python.org/3/c-api/long.html
Tying this back to our session processor, we were using `is` to compare the queue item ids for equality. Our queue item ids start at 0, and each queue item created increments this by one. So this comparison works only for the first 256 queue items on the machine.
Starting with the 257th queue item, the comparison starts returning `False`, and cancelation gets weird.
Easy fix - use `!=` instead of `is not`.
2024-12-02 23:22:58 -05:00
Mary Hipp
a6be26e174
fix(worker): only apply processor cancel logic if cancel event is for current queue item
2024-12-02 14:03:05 -08:00
Riku
159b00a490
fix(app): adjust session queue api type
2024-12-01 20:06:05 -08:00
Jonathan
6012b0f912
Update flux_text_encoder.py
...
Updated version number for FLUX Text Encoding.
2024-11-30 08:29:21 -05:00
Jonathan
bb0ed5dc8a
Update flux_denoise.py
...
Updated node version for FLUX Denoise.
2024-11-30 08:29:21 -05:00
psychedelicious
5910892c33
Merge remote-tracking branch 'origin/main' into ryan/flux-regional-prompting
2024-11-29 15:19:39 +10:00
Ryan Dick
8cfb032051
Add utility ImagePanelLayoutInvocation for working with In-Context LoRA workflows.
2024-11-26 20:58:31 -08:00
Ryan Dick
06a9d4e2b2
Use a Textarea component for the FluxTextEncoderInvocation prompt field.
2024-11-26 20:58:31 -08:00
Brandon Rising
ed46acee79
fix: Fail scan on InvalidMagicError in picklescan, update default for read_checkpoint_meta to scan unless explicitly told not to
2024-11-26 16:17:12 -05:00
Ryan Dick
53abdde242
Update Flux RegionalPromptingExtension to prepare both a mask with restricted image self-attention and a mask with unrestricted image self attention.
2024-11-25 22:04:23 +00:00
Ryan Dick
3741a6f5e0
Fix device handling for regional masks and apply the attention mask in the FLUX double stream block.
2024-11-25 16:02:03 +00:00
Ryan Dick
20356c0746
Fixup the logic for preparing FLUX regional prompt attention masks.
2024-11-21 22:46:25 +00:00
Ryan Dick
fda7aaa7ca
Pass RegionalPromptingExtension down to the CustomDoubleStreamBlockProcessor in FLUX.
2024-11-20 19:48:04 +00:00
Ryan Dick
85c616fa34
WIP - Pass prompt masks to FLUX model during denoising.
2024-11-20 18:51:43 +00:00
psychedelicious
f01210861b
chore: ruff
2024-11-19 07:02:37 -08:00
psychedelicious
872a6ef209
tidy(nodes): extract slerp from lblend to util fn
2024-11-19 07:02:37 -08:00
psychedelicious
4267e5ffc4
tidy(nodes): bring masked blend latents masking logic into invoke core
2024-11-19 07:02:37 -08:00
Brandon Rising
1fd80d54a4
Run Ruff
2024-11-19 07:02:37 -08:00
