- 
                Notifications
    You must be signed in to change notification settings 
- Fork 1.5k
Description
Traceback (most recent call last): File "E:\FramePack.git\webui\demo_gradio.py", line 241, in worker generated_latents = sample_hunyuan( File "E:\FramePack.git\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context return func(*args, **kwargs) File "E:\FramePack.git\webui\diffusers_helper\pipelines\k_diffusion_hunyuan.py", line 116, in sample_hunyuan results = sample_unipc(k_model, latents, sigmas, extra_args=sampler_kwargs, disable=False, callback=callback) File "E:\FramePack.git\webui\diffusers_helper\k_diffusion\uni_pc_fm.py", line 141, in sample_unipc return FlowMatchUniPC(model, extra_args=extra_args, variant=variant).sample(noise, sigmas=sigmas, callback=callback, disable_pbar=disable) File "E:\FramePack.git\webui\diffusers_helper\k_diffusion\uni_pc_fm.py", line 118, in sample model_prev_list = [self.model_fn(x, vec_t)] File "E:\FramePack.git\webui\diffusers_helper\k_diffusion\uni_pc_fm.py", line 23, in model_fn return self.model(x, t, **self.extra_args) File "E:\FramePack.git\webui\diffusers_helper\k_diffusion\wrapper.py", line 37, in k_model pred_positive = transformer(hidden_states=hidden_states, timestep=timestep, return_dict=False, **extra_args['positive'])[0].float() File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl return forward_call(*args, **kwargs) File "E:\FramePack.git\webui\diffusers_helper\models\hunyuan_video_packed.py", line 973, in forward hidden_states, encoder_hidden_states = self.gradient_checkpointing_method( File "E:\FramePack.git\webui\diffusers_helper\models\hunyuan_video_packed.py", line 832, in gradient_checkpointing_method result = block(*args) File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl return forward_call(*args, **kwargs) File "E:\FramePack.git\webui\diffusers_helper\models\hunyuan_video_packed.py", line 652, in forward attn_output, context_attn_output = self.attn( File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl return forward_call(*args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\diffusers\models\attention_processor.py", line 605, in forward return self.processor( File "E:\FramePack.git\webui\diffusers_helper\models\hunyuan_video_packed.py", line 172, in __call__ hidden_states = attn_varlen_func(query, key, value, cu_seqlens_q, cu_seqlens_kv, max_seqlen_q, max_seqlen_kv) File "E:\FramePack.git\webui\diffusers_helper\models\hunyuan_video_packed.py", line 111, in attn_varlen_func x = sageattn(q, k, v, tensor_layout='NHD') File "E:\FramePack.git\system\python\lib\site-packages\sageattention\core.py", line 132, in sageattn return sageattn_qk_int8_pv_fp8_cuda(q, k, v, tensor_layout=tensor_layout, is_causal=is_causal, sm_scale=sm_scale, return_lse=return_lse, pv_accum_dtype="fp32+fp32") File "E:\FramePack.git\system\python\lib\site-packages\torch\_dynamo\eval_frame.py", line 745, in _fn return fn(*args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\sageattention\core.py", line 720, in sageattn_qk_int8_pv_fp8_cuda q_int8, q_scale, k_int8, k_scale = per_thread_int8_triton(q, k, km, tensor_layout=tensor_layout, BLKQ=128, WARPQ=32, BLKK=64, WARPK=64) File "E:\FramePack.git\system\python\lib\site-packages\sageattention\triton\quant_per_thread.py", line 187, in per_thread_int8 quant_query_per_thread_int8_kernel[grid]( File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\jit.py", line 330, in <lambda> return lambda *args, **kwargs: self.run(grid=grid, warmup=False, *args, **kwargs) File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\jit.py", line 568, in run device = driver.active.get_current_device() File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\driver.py", line 23, in __getattr__ self._initialize_obj() File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\driver.py", line 20, in _initialize_obj self._obj = self._init_fn() File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\driver.py", line 9, in _create_driver return actives[0]() File "E:\FramePack.git\system\python\lib\site-packages\triton\backends\nvidia\driver.py", line 499, in __init__ self.utils = CudaUtils()  # TODO: make static File "E:\FramePack.git\system\python\lib\site-packages\triton\backends\nvidia\driver.py", line 96, in __init__ mod = compile_module_from_src(Path(os.path.join(dirname, "driver.c")).read_text(), "cuda_utils") File "E:\FramePack.git\system\python\lib\site-packages\triton\backends\nvidia\driver.py", line 58, in compile_module_from_src cache = get_cache_manager(key) File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\cache.py", line 285, in get_cache_manager return __cache_cls(_base64(key)) File "E:\FramePack.git\system\python\lib\site-packages\triton\runtime\cache.py", line 69, in __init__ os.makedirs(self.cache_dir, exist_ok=True) File "os.py", line 215, in makedirs File "os.py", line 215, in makedirs File "os.py", line 225, in makedirs PermissionError: [WinError 5] Access is denied: 'C:\\Users\\user\\.triton'
I have to clarify its been a few years since my uni year and I pretty much illiterate in setups.
I followed this guide as reference:
https://www.reddit.com/r/StableDiffusion/comments/1k34bot/installing_xformers_triton_flashsage_attention_on/
All 3 xformers, flash and sage is installed, but the problem seems to lie with triton which is required by sage.
Yea I have no idea sorry.