Commit History

Replace LTX enhancer with Sulphur GGUF enhancer
b3586c1

daKhosa commited on

Use Sulphur LoRAs and strengthen prompt handling
1efc965

daKhosa commited on

Load raw Gemma weights with Sulphur key mapping
83e482c

daKhosa commited on

Defer LTX model loading to GPU worker
0b97720

daKhosa commited on

Alias quantized Gemma weights for LTX loader
0a04b27

daKhosa commited on

Align LTX Space runtime dependencies
050a9f8

daKhosa commited on

Avoid xformers Dynamo import in LTX app
41f5b1e

daKhosa commited on

Avoid torch dynamo import under ZeroGPU
1bd58d5

daKhosa commited on

Set Sulphur LoRA strength to one
f430017

daKhosa commited on

Replace Wan2GP wrapper with direct LTX pipeline
2828873

daKhosa commited on

Stream WanGP events and cache uploads
cb6f99f

daKhosa commited on

Remove ZeroGPU duration and log WanGP startup
1efde5f

daKhosa commited on

Prefetch LTX distilled LoRA at startup
3831078

daKhosa commited on

Prefetch WanGP runtime assets at startup
f1be1c7

daKhosa commited on

Fix WanGP LTX model selection
c291abc

daKhosa commited on

Fix ZeroGPU startup CUDA initialization
00b47d9

daKhosa commited on

Warm WanGP session at startup
2944b5b

daKhosa commited on

Capture stderr (tqdm/progress) in log stream alongside stdout
f5b01ae

daKhosa commited on

Persist WanGPSession across generations; eliminate per-call model reload
deab82a

daKhosa commited on

Replace Frames slider with Seconds (1-5s); convert to frames internally at 25fps
36ea2bf

daKhosa commited on

Single-quote 'Experimental' in subtitle
3c276bc

daKhosa commited on

Update subtitle to reflect experimental LoRA v1
1776eeb

daKhosa commited on

Pre-download ltx-2.3-22b-distilled-lora-384 required by experimental LoRA
8d83ec8

daKhosa commited on

Switch to experimental Sulphur LoRA v1
9d642bd

daKhosa commited on

Untrack 10Eros subfolder; ignore it from Sulphur repo
ab07e2b

daKhosa commited on

Add 10Eros Space (TenStrip/LTX2.3-10Eros layer-scaled merge)
4bd6143

daKhosa commited on

Pre-download Gemma at startup; bump guidance to 5.0 for prompt adherence
499128d

Daankular commited on

Reduce GPU duration to 120s (H200)
ceb42ce

Daankular commited on

Flatten hf_hub_download subfolder so LoRA lands where Wan2GP expects it
0103a70

Daankular commited on

Use correct Sulphur distill LoRA from SulphurAI/Sulphur-2-base/distill_loras/
483e424

Daankular commited on

Stream tqdm progress lines by reading chunks and splitting on \r
e131563

Daankular commited on

Stream subprocess output line-by-line; increase GPU duration to 360s
24a1162

Daankular commited on

Fix: don't use rank-768 LoRA alongside full distil model; capture stderr
4fe229d

Daankular commited on

Scan output dir for video if artifacts list is empty
031028a

Daankular commited on

Use generate.py subprocess approach like the original β€” no direct WanGP imports in app
c838058

Daankular commited on

Add smplfitter and chumpy wheels
1f236a2

Daankular commited on

Remove gradio_rangeslider (requires gradio<6, conflicts)
cf7d1a6

Daankular commited on

All deps in requirements.txt; remove all runtime pip install and patch logic
f4ac1b9

Daankular commited on

Patch Wan2GP source to make rembg import optional after clone
acd62f0

Daankular commited on

Install Wan2GP deps via filtered requirements in setup() β€” no more whack-a-mole
051110a

Daankular commited on

Add all Wan2GP core deps to requirements.txt at once
8ef80ba

Daankular commited on

Add diffusers to requirements
b91b746

Daankular commited on

Fix Gradio launch: remove module-level chdir, skip Wan2GP pip install
ad05bf8

Daankular commited on

Fix colorFrom to valid HF value
d6487bb

Daankular commited on

Resolve README conflict, use sdk_version 6.14.0
a1fe250

Daankular commited on

Initial Sulphur-2 image-to-video Gradio Space
7469732

Daankular commited on

initial commit
db3c96f
verified

Daankular commited on