Wagner Bruna
bc762fe9b4
add support for Qwen Image Pruning ( #1779 )
...
From leejet/stable-diffusion.cpp#874 .
2025-10-10 16:22:47 +08:00
Wagner Bruna
bece22f996
fix encoding VAE tiling for Qwen Image ( #1785 )
2025-10-10 10:07:50 +08:00
Wagner Bruna
c48999f7c0
additional options for image generation ( #1765 )
...
* sd: add backend support for choosing the default sampler
* use the default sampler on the API
* sd: add backend support for the scheduler
* sd: add backend support for distilled guidance
* sd: add backend support for timestep-shift
* sd: add a config field to set default image gen options
2025-10-05 23:36:20 +08:00
Concedo
75272f62af
remove gif-h
2025-10-05 17:49:29 +08:00
Concedo
a98b63013e
allow tiling on qwen image
2025-10-04 20:43:36 +08:00
Concedo
bb06956b2d
allow wan to use img2img via init image
2025-10-04 11:25:46 +08:00
Concedo
db37688b47
qwen image disable VAE tiling as it's broken
2025-10-04 11:19:19 +08:00
Concedo
7447a362d1
hide a debug print
2025-10-03 20:56:57 +08:00
Concedo
15249baea1
apply jeffbolz f32 patch https://github.com/leejet/stable-diffusion.cpp/pull/851#issuecomment-3335515302
2025-10-03 19:18:46 +08:00
Concedo
f282362414
added qwen image support (+1 squashed commits)
...
Squashed commits:
[92df28061] added qwen image support (+1 squashed commits)
Squashed commits:
[1485c71ed] wip adding qwen image
2025-10-03 18:58:48 +08:00
Concedo
4f8f0e5949
move embeds into their own dir, detach sd vocab into separate files
2025-10-03 14:21:09 +08:00
Concedo
f2ad0b78d4
refactor of added images done
2025-10-02 21:22:35 +08:00
Concedo
df87da4694
wip refactor ref image bufs
2025-10-02 16:37:17 +08:00
Wagner Bruna
ac6be8ab8c
sd: do not force T5 on CPU anymore ( #1769 )
...
We now have the clip_cpu config parameter for that. Todo: Will make clip cpu on by default
2025-10-02 10:48:16 +08:00
Concedo
4587ccb71a
prepare to refactor reference image
2025-10-02 10:41:29 +08:00
Concedo
e49ac6b120
allow clip_vision to be loaded via clip_l or clip_g param
2025-10-01 17:57:49 +08:00
Concedo
2fc31d36c0
gif mime type for animated images
2025-10-01 17:18:00 +08:00
Concedo
9e4c29fda7
generate both gif and pick smaller (+1 squashed commits)
...
Squashed commits:
[09122d052] generate both gif and pick the smaller one
2025-09-30 14:42:58 +08:00
Concedo
4117542eae
switch to msf gif
2025-09-30 13:56:41 +08:00
Concedo
4b1c89ca5c
can save animated gifs
2025-09-29 22:52:42 +08:00
Concedo
7b71742223
added additional toggles for SD, sdoffloadcpu , sdvaecpu and sdclipcpu
2025-09-28 22:08:29 +08:00
Concedo
553890f975
compact UI launcher, WIP handling video outputs
2025-09-28 17:24:04 +08:00
Concedo
cd6c771bd9
WAN single images working
2025-09-28 15:01:19 +08:00
Concedo
cbb8a3feb5
taesd disable if not valid model arch
2025-09-27 17:00:30 +08:00
Wagner Bruna
42087c3622
update stable-diffusion.cpp to master-306-2abe945 ( #1732 )
...
* update stable-diffusion.cpp to master-52a97b3
* update stable-diffusion.cpp to master-0ebe6fe
* update stable-diffusion.cpp to master-301-fd693ac
* update stable-diffusion.cpp to master-306-2abe945
* fix taesd file selection
2025-09-27 16:52:58 +08:00
Wagner Bruna
01c1acc9a2
sdtype_adapter refactor ( #1753 )
...
* reorganize sampler selection in sdtype_adapter
* simplify image metadata generation
Use a stringstream to avoid the explicit conversions to
string. Also, generate from the sd.cpp struct, to reduce
dependencies on the global context fields.
* clean up default parameters in sdtype_adapter
Most image generation parameters are not customizable through
the API, and their values simply repeat the default set by
sd.cpp. So, remove them from the SDParams struct, and place
the needed constant values directly on the code.
Also, print debug information directly from the sd.cpp
parameters. And to help with type mismatches and general
readability, replace the fromatting with stringstreams.
2025-09-24 22:09:17 +08:00
Concedo
c7a1eec4e4
try to solve ttscpp oom regression
2025-09-24 17:45:28 +08:00
Concedo
84a67d289c
tts show model filename
2025-09-23 22:42:43 +08:00
Concedo
59b6a09ae1
try to fix kokoro alloc again
2025-09-22 21:22:41 +08:00
Concedo
13bee0d39d
some minor fixes
2025-09-22 13:20:06 +08:00
Concedo
8018e5222a
try fix compile issues rocm (+4 squashed commit)
...
Squashed commit:
[9e33a33f2] try fix compile issues rocm
[36e5eb56b] try fix compile issues rocm
[6ca18812b] try fix compile issues rocm
[b56ad2e8f] try fix compile issues rocm
2025-09-21 22:31:09 +08:00
Wagner Bruna
dd0bf706a3
fix Chroma workaround for flash attention ( #1746 )
...
chroma_use_dit_mask is a context parameter, so changing it
after creating the context has no effect.
2025-09-21 20:53:40 +08:00
Concedo
a5580a32fb
fix cuda and macos compile issues
2025-09-12 20:53:42 +08:00
Wagner Bruna
73be216062
sync stb_image_write.h with the sd.cpp version ( #1733 )
...
This is mainly to avoid keeping track of the small diff in
otherarch/sdcpp/main.cpp .
2025-09-11 20:37:35 +08:00
Concedo
7e35954695
Merge branch 'upstream' into concedo_experimental
...
# Conflicts:
# docs/build.md
# docs/function-calling.md
# examples/eval-callback/eval-callback.cpp
# ggml/CMakeLists.txt
# ggml/src/ggml-cann/ggml-cann.cpp
# ggml/src/ggml-cpu/CMakeLists.txt
# ggml/src/ggml-cpu/kleidiai/kernels.cpp
# ggml/src/ggml-cpu/kleidiai/kernels.h
# ggml/src/ggml-cpu/kleidiai/kleidiai.cpp
# scripts/compare-llama-bench.py
# scripts/server-bench.py
# scripts/tool_bench.py
# tests/test-chat.cpp
# tools/batched-bench/batched-bench.cpp
# tools/llama-bench/llama-bench.cpp
# tools/server/README.md
2025-08-31 23:33:36 +08:00
Concedo
7b396bd917
added v1 voices endpoint, added lcpp aliases for cli, fixed dia wrong voice
2025-08-30 11:20:18 +08:00
Concedo
75c919cfd4
can't resolve the clicking
2025-08-26 17:55:03 +08:00
Concedo
ff1d179e21
kokoro fixes for every better pausing
2025-08-25 22:28:50 +08:00
Concedo
205a0b8d4c
fix kokoro replacement, add 4096 batch size option
2025-08-25 15:57:13 +08:00
Wagner Bruna
f41fc2579e
fix ttscppmain build ( #1707 )
2025-08-25 10:22:33 +08:00
Concedo
02e05f9321
fix for comma pausing in kokoro
2025-08-25 00:18:00 +08:00
Concedo
b0a8d11584
add tts max length for kokoro (+1 squashed commits)
...
Squashed commits:
[c1c6feaf] add tts max length for kokoro
2025-08-24 17:57:29 +08:00
Concedo
3867db34bc
adjust kokoro default voices
Copilot Setup Steps / copilot-setup-steps (push) Has been cancelled
2025-08-22 23:48:29 +08:00
Concedo
80dabbb689
minor adjustments for sdquant: allow backend to do the translation for the type more defensively, adjust the UI dropdown for clarity.
2025-08-22 23:23:32 +08:00
Wagner Bruna
2f8b0ec538
Support q8_0 quantization for image model loading ( #1692 )
...
* Support q8_0 quantization for image model loading
q4_0 may degrade quality significantly, especially for smaller
models like SD 1.5 and SDXL. q8_0 provides a middle-ground,
giving half the memory savings of q4_0 but loading faster and
with less quality loss.
* Accept --sdquant with no parameters
* Use numerical values for the sdquant option
2025-08-22 22:17:15 +08:00
Concedo
9dd6b4c930
improve whisper transcribe apt regex
2025-08-22 17:13:51 +08:00
Wagner Bruna
6003e90e50
Add flash attention and conv2d direct controls for image generation ( #1678 )
...
* Add separate flash attention config for image generation
* Add config option for Conv2D Direct
2025-08-20 12:17:57 +08:00
Concedo
56c0926866
fix metal build
2025-08-19 15:41:16 +08:00
Concedo
9fb0611115
handle contractions correctly, bump defaults
2025-08-18 22:33:44 +08:00
Concedo
2abe11071b
custom voice handling
2025-08-18 16:57:34 +08:00
