Error after loading weights in pytorch #39

Hi

I have problem in generation, could anybody help? :) I'm new to this stuff but want to test it :) I'm running i7-12700k with 4090

ython demo.py --low_vram -p "A beautiful landscape with a river and mountains"
Detected GPU VRAM less than 24GB, setting low_vram to True
2026-04-15 16:06:33,531 - INFO - server listening on 0.0.0.0:8080
╭──────────────── viser ────────────────╮
│ ╷ │
│ HTTP │ http://localhost:8080 │
│ Websocket │ ws://localhost:8080 │
│ ╵ │
╰───────────────────────────────────────╯
2026-04-15 16:06:33,723 - INFO - HTTP Request: HEAD https://huggingface.co/haodongli/DA-2/resolve/main/config.json "HTTP/1.1 404 Not Found"
2026-04-15 16:06:34,841 - INFO - HTTP Request: HEAD https://huggingface.co/haodongli/DA-2/resolve/main/model.safetensors "HTTP/1.1 302 Found"
2026-04-15 16:06:35,436 - INFO - HTTP Request: HEAD https://huggingface.co/LeoXie/WorldGen/resolve/main/models--WorldGen-Flux-Lora/worldgen_text2scene.safetensors "HTTP/1.1 302 Found"
Using Nunchaku with int4 precision
/home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/huggingface_hub/utils/_validators.py:205: UserWarning: The local_dir_use_symlinks argument is deprecated and ignored in hf_hub_download. Downloading to a local directory does not use symlinks anymore.
warnings.warn(
2026-04-15 16:06:35,588 - INFO - HTTP Request: HEAD https://huggingface.co/mit-han-lab/svdq-int4-flux.1-dev/resolve/main/unquantized_layers.safetensors "HTTP/1.1 302 Found"
2026-04-15 16:06:35,745 - INFO - HTTP Request: HEAD https://huggingface.co/mit-han-lab/svdq-int4-flux.1-dev/resolve/main/transformer_blocks.safetensors "HTTP/1.1 302 Found"
2026-04-15 16:06:35,891 - INFO - HTTP Request: HEAD https://huggingface.co/mit-han-lab/svdq-int4-flux.1-dev/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
2026-04-15 16:06:35,910 - INFO - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/mit-han-lab/svdq-int4-flux.1-dev/0c824f154235c9fbddff83e8c495d833217cd76c/config.json "HTTP/1.1 200 OK"
[2026-04-15 16:06:36.131] [info] Initializing QuantizedFluxModel on device 0
[2026-04-15 16:06:36.131] [info] Layer offloading enabled
[2026-04-15 16:06:36.140] [info] Loading weights from /home/fox/.cache/huggingface/hub/models--mit-han-lab--svdq-int4-flux.1-dev/snapshots/0c824f154235c9fbddff83e8c495d833217cd76c/transformer_blocks.safetensors
[2026-04-15 16:06:37.674] [info] Done.
Injecting quantized module
2026-04-15 16:06:37,844 - INFO - HTTP Request: GET https://huggingface.co/api/models/black-forest-labs/FLUX.1-dev "HTTP/1.1 200 OK"
2026-04-15 16:06:37,997 - INFO - HTTP Request: HEAD https://huggingface.co/black-forest-labs/FLUX.1-dev/resolve/main/model_index.json "HTTP/1.1 200 OK"
Keyword arguments {'device': device(type='cuda')} are not expected by FluxPipeline and will be ignored.
Loading weights: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 219/219 [00:00<00:00, 15590.36it/s]
Loading weights: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 196/196 [00:00<00:00, 15598.72it/s]
Loading pipeline components...: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 35.14it/s]
Loading LoRA weights from: /home/fox/.cache/huggingface/hub/models--LeoXie--WorldGen/snapshots/2e2b2cb4dc3144a65481f0ca26b892584436868c/models--WorldGen-Flux-Lora/worldgen_text2scene.safetensors
Converting LoRAs to nunchaku format: 0%| | 0/57 [00:00<?, ?it/s]/home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/lora/flux/utils.py:43: UserWarning: Using a non-tuple sequence for multidimensional indexing is deprecated and will be changed in pytorch 2.9; use x[tuple(seq)] instead of x[seq]. In pytorch 2.9 this will be interpreted as tensor index, x[torch.tensor(seq)], which will result either in an error or a different result (Triggered internally at /pytorch/torch/csrc/autograd/python_variable_indexing.cpp:347.)
result[[slice(0, extent) for extent in tensor.shape]] = tensor
Converting LoRAs to nunchaku format: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 57/57 [00:00<00:00, 140.66it/s]
[2026-04-15 16:06:38.662] [info] Loading partial weights from pytorch
Traceback (most recent call last):
File "/home/fox/Downloads/WorldGen/demo.py", line 366, in
server = ViserServer(args)
^^^^^^^^^^^^^^^^^
File "/home/fox/Downloads/WorldGen/demo.py", line 68, in init
self.worldgen = WorldGen(mode=mode,
^^^^^^^^^^^^^^^^^^^
File "/home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/worldgen/worldgen.py", line 38, in init
self.pano_gen_model = build_pano_gen_model(lora_path=lora_path, device=device, low_vram=low_vram)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/worldgen/pano_gen.py", line 37, in build_pano_gen_model
transformer.update_lora_params(state_dict)
File "/home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/models/transformers/transformer_flux.py", line 459, in update_lora_params
block.m.loadDict(state_dict, True)
RuntimeError: Cannot access data pointer of Tensor that doesn't have storage
Exception raised from throw_data_ptr_access_error at /pytorch/c10/core/TensorImpl.cpp:307 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits, std::allocator >) + 0x9d (0x7bd39157305d in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const, char const, unsigned int, char const*) + 0x69 (0x7bd3915077d3 in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #2: c10::TensorImpl::throw_data_ptr_access_error() const + 0x34 (0x7bd39154c634 in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #3: + 0x59d37f (0x7bd178f9d37f in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #4: + 0x59baf6 (0x7bd178f9baf6 in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #5: + 0x51adbd (0x7bd178f1adbd in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #6: + 0x5110ae (0x7bd178f110ae in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #7: + 0x510e9b (0x7bd178f10e9b in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #8: + 0x510e9b (0x7bd178f10e9b in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #9: + 0x510e9b (0x7bd178f10e9b in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #10: + 0x54ddef (0x7bd178f4ddef in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #11: + 0x542797 (0x7bd178f42797 in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #12: + 0x54285f (0x7bd178f4285f in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)
frame #13: + 0x538310 (0x7bd178f38310 in /home/fox/anaconda3/envs/worldgen/lib/python3.11/site-packages/nunchaku/_C.cpython-311-x86_64-linux-gnu.so)

frame #34: + 0x2a1ca (0x7bd3ca42a1ca in /lib/x86_64-linux-gnu/libc.so.6)
frame #35: __libc_start_main + 0x8b (0x7bd3ca42a28b in /lib/x86_64-linux-gnu/libc.so.6)
**