☑️ Так же вы можете закинуть например позу из openpose и попросить поставить персонажа в эту позу

☑️ Улучшенная согласованность редактирования текста на английском и китайском

😱 Но и это еще не все.

Теперь мы можем работать в разрешении от 1 до 2 мегапикселей❗️ что улучшает детализацию, которая в первой Qwen Edit сильно хромала.

🚀Уже вышла Nunchaku.

🎦 СМОТРЕТЬ НА YOUTUBE (https://youtu.be/hJzjQ1XhAfw)

🎦 СМОТРЕТЬ НА RUTUBE (https://rutube.ru/video/97a51b1d0ae3eebc274fc05e76136872/)

Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug 1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]

Version: f2.0.1v1.10.1-previous-669-gdfdcbab6

Commit hash: dfdcbab685e57677014f05a3309b48cc87383167

Launching Web UI with arguments: --skip-torch-cuda-test --no-download-sd-model

D:\SD\system\python\lib\site-packages\torch\cuda\__init__.py:209: UserWarning:

NVIDIA GeForce RTX 5060 Laptop GPU with CUDA capability sm_120 is not compatible with the current PyTorch installation.

The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90.

If you want to use the NVIDIA GeForce RTX 5060 Laptop GPU GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

warnings.warn(

Total VRAM 8151 MB, total RAM 16084 MB

pytorch version: 2.3.1+cu121

Set vram state to: NORMAL_VRAM

Device: cuda:0 NVIDIA GeForce RTX 5060 Laptop GPU : native

Hint: your device supports --cuda-malloc for potential speed improvements.

VAE dtype preferences: [torch.bfloat16, torch.float32] -> torch.bfloat16

CUDA Using Stream: False

D:\SD\system\python\lib\site-packages\transformers\utils\hub.py:128: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.

warnings.warn(

Using pytorch cross attention

Using pytorch attention for VAE

ControlNet preprocessor location: D:\SD\webui\models\ControlNetPreprocessor

2025-09-28 09:52:00,089 - ControlNet - INFO - ControlNet UI callback registered.

Model selected: {'checkpoint_info': {'filename': 'D:\\SD\\webui\\models\\Stable-diffusion\\gonzalomoXLFluxPony_v40UnityXLDMD.safetensors', 'hash': 'd632a450'}, 'additional_modules': [], 'unet_storage_dtype': None}

Using online LoRAs in FP16: False

Running on local URL: http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.

Startup time: 57.7s (prepare environment: 0.7s, launcher: 0.4s, import torch: 25.1s, initialize shared: 0.4s, other imports: 1.5s, load scripts: 8.5s, create ui: 13.8s, gradio launch: 7.3s).

Environment vars changed: {'stream': False, 'inference_memory': 1024.0, 'pin_shared_memory': False}

[GPU Setting] You will use 87.44% GPU memory (7126.00 MB) to load weights, and use 12.56% GPU memory (1024.00 MB) to do matrix computation.

Loading Model: {'checkpoint_info': {'filename': 'D:\\SD\\webui\\models\\Stable-diffusion\\gonzalomoXLFluxPony_v40UnityXLDMD.safetensors', 'hash': 'd632a450'}, 'additional_modules': [], 'unet_storage_dtype': None}

[Unload] Trying to free all memory for cuda:0 with 0 models keep loaded ... Done.

StateDict Keys: {'unet': 1680, 'vae': 248, 'text_encoder': 197, 'text_encoder_2': 518, 'ignore': 0}

Working with z of shape (1, 4, 32, 32) = 4096 dimensions.

K-Model Created: {'storage_dtype': torch.float16, 'computation_dtype': torch.float16}

Model loaded in 22.0s (unload existing model: 0.6s, forge model load: 21.5s).

Skipping unconditional conditioning when CFG = 1. Negative Prompts are ignored.

[Unload] Trying to free 3051.58 MB for cuda:0 with 0 models keep loaded ... Done.

[Memory Management] Target: JointTextEncoder, Free GPU: 1934.60 MB, Model Require: 1559.68 MB, Previously Loaded: 0.00 MB, Inference Require: 1024.00 MB, Remaining: -649.08 MB, CPU Swap Loaded (blocked method): 1056.00 MB, GPU Loaded: 696.68 MB

Moving model(s) has taken 3.21 seconds

Traceback (most recent call last):

File "D:\SD\webui\modules_forge\main_thread.py", line 30, in work

self.result = self.func(*self.args, **self.kwargs)

File "D:\SD\webui\modules\txt2img.py", line 131, in txt2img_function

processed = processing.process_images(p)

File "D:\SD\webui\modules\processing.py", line 842, in process_images

res = process_images_inner(p)

File "D:\SD\webui\modules\processing.py", line 962, in process_images_inner

p.setup_conds()

File "D:\SD\webui\modules\processing.py", line 1601, in setup_conds

super().setup_conds()

File "D:\SD\webui\modules\processing.py", line 505, in setup_conds

self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, total_steps, [self.cached_c], self.extra_network_data)

File "D:\SD\webui\modules\processing.py", line 474, in get_conds_with_caching

cache[1] = function(shared.sd_model, required_prompts, steps, hires_steps, shared.opts.use_old_scheduling)

File "D:\SD\webui\modules\prompt_parser.py", line 262, in get_multicond_learned_conditioning

learned_conditioning = get_learned_conditioning(model, prompt_flat_list, steps, hires_steps, use_old_scheduling)

File "D:\SD\webui\modules\prompt_parser.py", line 189, in get_learned_conditioning

conds = model.get_learned_conditioning(texts)

File "D:\SD\system\python\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context

return func(*args, **kwargs)

File "D:\SD\webui\backend\diffusion_engine\sdxl.py", line 89, in get_learned_conditioning

cond_l = self.text_processing_engine_l(prompt)

File "D:\SD\webui\backend\text_processing\classic_engine.py", line 272, in __call__

z = self.process_tokens(tokens, multipliers)

File "D:\SD\webui\backend\text_processing\classic_engine.py", line 305, in process_tokens

z = self.encode_with_transformers(tokens)

File "D:\SD\webui\backend\text_processing\classic_engine.py", line 128, in encode_with_transformers

self.text_encoder.transformer.text_model.embeddings.position_embedding = self.text_encoder.transformer.text_model.embeddings.p...(dtype=torch.float32)

File "D:\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1173, in to

return self._apply(convert)

File "D:\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 804, in _apply

param_applied = fn(param)

File "D:\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1159, in convert

return t.to(

RuntimeError: CUDA error: no kernel image is available for execution on the device

CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.

For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

CUDA error: no kernel image is available for execution on the device

CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.

For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

Я не научилась пользоваться Питоном сразу как это вошло в моду, но благо, научилась пользоваться переводчиком в далеком 2011.
Примерно понимаю, что какой-то CUDA (туда) несовместим с Пэйторч. (который, к слову, я через пип качала неоднократно зачем-то) Я вообще мало понимаю что и зачем я делала.
По ссылке я переходила и там качала пэйторч. Или что-то качала. В основном не интуитивно, а по каким-то там гайдам, на авось надеясь, что проблема решится.

В самой диффузии при генерации изображения:

RuntimeError: CUDA error: no kernel image is available for execution on the device CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

Ниже приведу инфу по ноуту