various fixes

This commit is contained in:
DeepBeepMeep 2025-09-27 15:14:55 +02:00
parent 79df3aae64
commit 5ce8fc3d53
8 changed files with 66 additions and 16 deletions

View File

@ -122,7 +122,9 @@ See full changelog: **[Changelog](docs/CHANGELOG.md)**
## 🚀 Quick Start ## 🚀 Quick Start
**One-click installation:** Get started instantly with [Pinokio App](https://pinokio.computer/) **One-click installation:**
- Get started instantly with [Pinokio App](https://pinokio.computer/)
- Use Redtash1 [One Click Install with Sage](https://github.com/Redtash1/Wan2GP-Windows-One-Click-Install-With-Sage)
**Manual installation:** **Manual installation:**
```bash ```bash
@ -136,8 +138,7 @@ pip install -r requirements.txt
**Run the application:** **Run the application:**
```bash ```bash
python wgp.py # Text-to-video (default) python wgp.py
python wgp.py --i2v # Image-to-video
``` ```
**Update the application:** **Update the application:**

View File

@ -232,7 +232,7 @@ class QwenImagePipeline(): #DiffusionPipeline
drop_idx = self.prompt_template_encode_start_idx drop_idx = self.prompt_template_encode_start_idx
txt = [template.format(e) for e in prompt] txt = [template.format(e) for e in prompt]
if self.processor is not None and image is not None: if self.processor is not None and image is not None and len(image) > 0:
img_prompt_template = "Picture {}: <|vision_start|><|image_pad|><|vision_end|>" img_prompt_template = "Picture {}: <|vision_start|><|image_pad|><|vision_end|>"
if isinstance(image, list): if isinstance(image, list):
base_img_prompt = "" base_img_prompt = ""
@ -972,7 +972,7 @@ class QwenImagePipeline(): #DiffusionPipeline
if callback is not None: if callback is not None:
preview = self._unpack_latents(latents, height, width, self.vae_scale_factor) preview = self._unpack_latents(latents, height, width, self.vae_scale_factor)
preview = preview.squeeze(0) preview = preview.transpose(0,2).squeeze(0)
callback(i, preview, False) callback(i, preview, False)

View File

@ -595,7 +595,7 @@ class WanAny2V:
color_reference_frame = input_frames[:, -1:].clone() color_reference_frame = input_frames[:, -1:].clone()
if prefix_frames_count > 0: if prefix_frames_count > 0:
overlapped_frames_num = prefix_frames_count overlapped_frames_num = prefix_frames_count
overlapped_latents_frames_num = (overlapped_latents_frames_num -1 // 4) + 1 overlapped_latents_frames_num = (overlapped_frames_num -1 // 4) + 1
# overlapped_latents_frames_num = overlapped_latents.shape[2] # overlapped_latents_frames_num = overlapped_latents.shape[2]
# overlapped_frames_num = (overlapped_latents_frames_num-1) * 4 + 1 # overlapped_frames_num = (overlapped_latents_frames_num-1) * 4 + 1
else: else:
@ -735,11 +735,20 @@ class WanAny2V:
if callback != None: if callback != None:
callback(-1, None, True) callback(-1, None, True)
clear_caches()
offload.shared_state["_chipmunk"] = False offload.shared_state["_chipmunk"] = False
chipmunk = offload.shared_state.get("_chipmunk", False) chipmunk = offload.shared_state.get("_chipmunk", False)
if chipmunk: if chipmunk:
self.model.setup_chipmunk() self.model.setup_chipmunk()
offload.shared_state["_radial"] = offload.shared_state["_attention"]=="radial"
radial = offload.shared_state.get("_radial", False)
if radial:
radial_cache = get_cache("radial")
from shared.radial_attention.attention import fill_radial_cache
fill_radial_cache(radial_cache, len(self.model.blocks), *target_shape[1:])
# init denoising # init denoising
updated_num_steps= len(timesteps) updated_num_steps= len(timesteps)

View File

@ -315,6 +315,8 @@ class WanSelfAttention(nn.Module):
x_ref_attn_map = None x_ref_attn_map = None
chipmunk = offload.shared_state.get("_chipmunk", False) chipmunk = offload.shared_state.get("_chipmunk", False)
radial = offload.shared_state.get("_radial", False)
if chipmunk and self.__class__ == WanSelfAttention: if chipmunk and self.__class__ == WanSelfAttention:
q = q.transpose(1,2) q = q.transpose(1,2)
k = k.transpose(1,2) k = k.transpose(1,2)
@ -322,12 +324,17 @@ class WanSelfAttention(nn.Module):
attn_layers = offload.shared_state["_chipmunk_layers"] attn_layers = offload.shared_state["_chipmunk_layers"]
x = attn_layers[self.block_no](q, k, v) x = attn_layers[self.block_no](q, k, v)
x = x.transpose(1,2) x = x.transpose(1,2)
elif radial and self.__class__ == WanSelfAttention:
qkv_list = [q,k,v]
del q,k,v
radial_cache = get_cache("radial")
no_step_no = offload.shared_state["step_no"]
x = radial_cache[self.block_no](qkv_list=qkv_list, timestep_no=no_step_no)
elif block_mask == None: elif block_mask == None:
qkv_list = [q,k,v] qkv_list = [q,k,v]
del q,k,v del q,k,v
x = pay_attention(
qkv_list, x = pay_attention( qkv_list, window_size=self.window_size)
window_size=self.window_size)
else: else:
with sdp_kernel(enable_flash=True, enable_math=False, enable_mem_efficient=False): with sdp_kernel(enable_flash=True, enable_math=False, enable_mem_efficient=False):
@ -1311,9 +1318,8 @@ class WanModel(ModelMixin, ConfigMixin):
del causal_mask del causal_mask
offload.shared_state["embed_sizes"] = grid_sizes offload.shared_state["embed_sizes"] = grid_sizes
offload.shared_state["step_no"] = real_step_no offload.shared_state["step_no"] = current_step_no
offload.shared_state["max_steps"] = max_steps offload.shared_state["max_steps"] = max_steps
if current_step_no == 0 and x_id == 0: clear_caches()
# arguments # arguments
kwargs = dict( kwargs = dict(

View File

@ -184,6 +184,15 @@ class family_handler():
"visible": False "visible": False
} }
# extra_model_def["image_ref_choices"] = {
# "choices": [("None", ""),
# ("People / Objects", "I"),
# ("Landscape followed by People / Objects (if any)", "KI"),
# ],
# "visible": False,
# "letters_filter": "KFI",
# }
extra_model_def["video_guide_outpainting"] = [0,1] extra_model_def["video_guide_outpainting"] = [0,1]
extra_model_def["keep_frames_video_guide_not_supported"] = True extra_model_def["keep_frames_video_guide_not_supported"] = True
extra_model_def["extract_guide_from_window_start"] = True extra_model_def["extract_guide_from_window_start"] = True

View File

@ -42,6 +42,11 @@ try:
except ImportError: except ImportError:
sageattn_varlen_wrapper = None sageattn_varlen_wrapper = None
try:
from spas_sage_attn import block_sparse_sage2_attn_cuda
except ImportError:
block_sparse_sage2_attn_cuda = None
try: try:
from .sage2_core import sageattn as sageattn2, is_sage2_supported from .sage2_core import sageattn as sageattn2, is_sage2_supported
@ -62,6 +67,8 @@ def sageattn2_wrapper(
return o return o
from sageattn import sageattn_blackwell as sageattn3
try: try:
from sageattn import sageattn_blackwell as sageattn3 from sageattn import sageattn_blackwell as sageattn3
except ImportError: except ImportError:
@ -144,6 +151,9 @@ def get_attention_modes():
ret.append("sage") ret.append("sage")
if sageattn2 != None and version("sageattention").startswith("2") : if sageattn2 != None and version("sageattention").startswith("2") :
ret.append("sage2") ret.append("sage2")
if block_sparse_sage2_attn_cuda != None and version("sageattention").startswith("2") :
ret.append("radial")
if sageattn3 != None: # and version("sageattention").startswith("3") : if sageattn3 != None: # and version("sageattention").startswith("3") :
ret.append("sage3") ret.append("sage3")
@ -159,6 +169,8 @@ def get_supported_attention_modes():
if not sage2_supported: if not sage2_supported:
if "sage2" in ret: if "sage2" in ret:
ret.remove("sage2") ret.remove("sage2")
if "radial" in ret:
ret.remove("radial")
if major < 7: if major < 7:
if "sage" in ret: if "sage" in ret:
@ -225,6 +237,7 @@ def pay_attention(
if attn == "chipmunk": if attn == "chipmunk":
from src.chipmunk.modules import SparseDiffMlp, SparseDiffAttn from src.chipmunk.modules import SparseDiffMlp, SparseDiffAttn
from src.chipmunk.util import LayerCounter, GLOBAL_CONFIG from src.chipmunk.util import LayerCounter, GLOBAL_CONFIG
if attn == "radial": attn ="sage2"
if b > 1 and k_lens != None and attn in ("sage2", "sage3", "sdpa"): if b > 1 and k_lens != None and attn in ("sage2", "sage3", "sdpa"):
assert attention_mask == None assert attention_mask == None

View File

@ -28,18 +28,24 @@ from sageattention.triton.quant_per_thread import per_thread_int8 as per_thread_
try: try:
from sageattention import _qattn_sm80 from sageattention import _qattn_sm80
if not hasattr(_qattn_sm80, "qk_int8_sv_f16_accum_f32_attn"):
_qattn_sm80 = torch.ops.sageattention_qattn_sm80
SM80_ENABLED = True SM80_ENABLED = True
except: except:
SM80_ENABLED = False SM80_ENABLED = False
try: try:
from sageattention import _qattn_sm89 from sageattention import _qattn_sm89
if not hasattr(_qattn_sm89, "qk_int8_sv_f8_accum_f32_fuse_v_scale_attn_inst_buf"):
_qattn_sm89 = torch.ops.sageattention_qattn_sm89
SM89_ENABLED = True SM89_ENABLED = True
except: except:
SM89_ENABLED = False SM89_ENABLED = False
try: try:
from sageattention import _qattn_sm90 from sageattention import _qattn_sm90
if not hasattr(_qattn_sm90, "qk_int8_sv_f8_accum_f32_fuse_v_scale_attn_inst_buf"):
_qattn_sm90 = torch.ops.sageattention_qattn_sm90
SM90_ENABLED = True SM90_ENABLED = True
except: except:
SM90_ENABLED = False SM90_ENABLED = False

16
wgp.py
View File

@ -63,7 +63,7 @@ AUTOSAVE_FILENAME = "queue.zip"
PROMPT_VARS_MAX = 10 PROMPT_VARS_MAX = 10
target_mmgp_version = "3.6.0" target_mmgp_version = "3.6.0"
WanGP_version = "8.74" WanGP_version = "8.75"
settings_version = 2.36 settings_version = 2.36
max_source_video_frames = 3000 max_source_video_frames = 3000
prompt_enhancer_image_caption_model, prompt_enhancer_image_caption_processor, prompt_enhancer_llm_model, prompt_enhancer_llm_tokenizer = None, None, None, None prompt_enhancer_image_caption_model, prompt_enhancer_image_caption_processor, prompt_enhancer_llm_model, prompt_enhancer_llm_tokenizer = None, None, None, None
@ -1394,6 +1394,12 @@ def _parse_args():
help="View form generation / refresh time" help="View form generation / refresh time"
) )
parser.add_argument(
"--betatest",
action="store_true",
help="test unreleased features"
)
parser.add_argument( parser.add_argument(
"--vram-safety-coefficient", "--vram-safety-coefficient",
type=float, type=float,
@ -4367,7 +4373,7 @@ def enhance_prompt(state, prompt, prompt_enhancer, multi_images_gen_type, overri
if image_start is None or not "I" in prompt_enhancer: if image_start is None or not "I" in prompt_enhancer:
image_start = [None] * num_prompts image_start = [None] * num_prompts
else: else:
image_start = [img[0] for img in image_start] image_start = [convert_image(img[0]) for img in image_start]
if len(image_start) == 1: if len(image_start) == 1:
image_start = image_start * num_prompts image_start = image_start * num_prompts
else: else:
@ -8712,9 +8718,9 @@ def generate_configuration_tab(state, blocks, header, model_family, model_choice
("Flash" + check("flash")+ ": good quality - requires additional install (usually complex to set up on Windows without WSL)", "flash"), ("Flash" + check("flash")+ ": good quality - requires additional install (usually complex to set up on Windows without WSL)", "flash"),
("Xformers" + check("xformers")+ ": good quality - requires additional install (usually complex, may consume less VRAM to set up on Windows without WSL)", "xformers"), ("Xformers" + check("xformers")+ ": good quality - requires additional install (usually complex, may consume less VRAM to set up on Windows without WSL)", "xformers"),
("Sage" + check("sage")+ ": 30% faster but slightly worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage"), ("Sage" + check("sage")+ ": 30% faster but slightly worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage"),
("Sage2/2++" + check("sage2")+ ": 40% faster but slightly worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage2"), ("Sage2/2++" + check("sage2")+ ": 40% faster but slightly worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage2")]\
("Sage3" + check("sage3")+ ": x2 faster but worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage3"), + ([("Radial" + check("radial")+ ": x? faster but ? quality - requires Sparge & Sage 2 Attn (usually complex to set up on Windows without WSL)", "radial")] if args.betatest else [])\
], + [("Sage3" + check("sage3")+ ": x2 faster but worse quality - requires additional install (usually complex to set up on Windows without WSL)", "sage3")],
value= attention_mode, value= attention_mode,
label="Attention Type", label="Attention Type",
interactive= not lock_ui_attention interactive= not lock_ui_attention