Unsupported CUDA architecture: sm60

My graphics cards are two Tesla P100 GPUs with CUDA 12.6. The following error occurred during operation. Please correct me, masters.

ERROR DURING GENERATION: Unsupported CUDA architecture: sm60

Traceback (most recent call last):
File "E:\AI\FramePack-V4\app.py", line 1223, in worker
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\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 "D:\AI-WORK\FramePack-V4\FramePack-V4\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 "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\k_diffusion\uni_pc_fm.py", line 118, in sample
model_prev_list = [self.model_fn(x, vec_t)]
File "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\k_diffusion\uni_pc_fm.py", line 23, in model_fn
return self.model(x, t, **self.extra_args)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\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 "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\models\hunyuan_video_packed.py", line 973, in forward
hidden_states, encoder_hidden_states = self.gradient_checkpointing_method(
File "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\models\hunyuan_video_packed.py", line 832, in gradient_checkpointing_method
result = block(*args)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\models\hunyuan_video_packed.py", line 652, in forward
attn_output, context_attn_output = self.attn(
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1739, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\torch\nn\modules\module.py", line 1750, in _call_impl
return forward_call(*args, **kwargs)
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\diffusers\models\attention_processor.py", line 605, in forward
return self.processor(
File "D:\AI-WORK\FramePack-V4\FramePack-V4\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 "D:\AI-WORK\FramePack-V4\FramePack-V4\diffusers_helper\models\hunyuan_video_packed.py", line 111, in attn_varlen_func
x = sageattn(q, k, v, tensor_layout='NHD')
File "D:\AI-WORK\FramePack-V4\FramePack-V4\deepface\lib\site-packages\sageattention\core.py", line 138, in sageattn
raise ValueError(f"Unsupported CUDA architecture: {arch}")
ValueError: Unsupported CUDA architecture: sm60

Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.

Start by reading diffusers_helper/models/hunyuan_video_packed.py at the sageattn call and deepface/lib/site-packages/sageattention/core.py where the sm60 error is raised. Check how the installed attention dependency handles the reported Tesla P100 architecture and whether the project documents supported hardware. Done should be a verified compatible path or a clear project-level explanation of the hardware limitation.