为何5070ti跑FramePack没有反应

hahaha253

星级打分

• 1
• 2
• 3
• 4
• 5

平均分:NAN  参与人数:0  我的评分:未评

出现一下生成文字。然后就变成这样了，各位老哥怎么解决的？

大概显示如下

Currently enabled native sdp backends: ['flash', 'math', 'mem_efficient', 'cudnn']
Xformers is not installed!
Flash Attn is not installed!
Sage Attn is not installed!
G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\system\python\lib\site-packages\torch\cuda\__init__.py:235: UserWarning:
NVIDIA GeForce RTX 5070 Ti 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 5070 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

  warnings.warn(
Namespace(share=False, server='127.0.0.1', port=7869, inbrowser=True)
Free VRAM 14.5732421875 GB
High-VRAM Mode: False
Downloading shards: 100%|██████████████████████████████████████████████████████████████| 4/4 [00:00<00:00, 2013.10it/s]
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████| 4/4 [00:01<00:00,  2.34it/s]
Fetching 3 files: 100%|██████████████████████████████████████████████████████████████████████████| 3/3 [00:00<?, ?it/s]
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████| 3/3 [00:03<00:00,  1.12s/it]
transformer.high_quality_fp32_output_for_inference = True
* Running on local URL:  http://127.0.0.1:7869

To create a public link, set `share=True` in `launch()`.
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Loaded CLIPTextModel to cuda:0 as complete.
Traceback (most recent call last):
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\demo_gradio.py", line 126, in worker
    llama_vec, clip_l_pooler = encode_prompt_conds(prompt, text_encoder, text_encoder_2, tokenizer, tokenizer_2)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\diffusers_helper\hunyuan.py", line 31, in encode_prompt_conds
    llama_attention_length = int(llama_attention_mask.sum())
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.

Unloaded CLIPTextModel as complete.
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Loaded CLIPTextModel to cuda:0 as complete.
Traceback (most recent call last):
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\demo_gradio.py", line 126, in worker
    llama_vec, clip_l_pooler = encode_prompt_conds(prompt, text_encoder, text_encoder_2, tokenizer, tokenizer_2)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\diffusers_helper\hunyuan.py", line 31, in encode_prompt_conds
    llama_attention_length = int(llama_attention_mask.sum())
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.

Unloaded CLIPTextModel as complete.
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Unloaded DynamicSwap_LlamaModel as complete.
Unloaded CLIPTextModel as complete.
Unloaded SiglipVisionModel as complete.
Unloaded AutoencoderKLHunyuanVideo as complete.
Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Loaded CLIPTextModel to cuda:0 as complete.
Traceback (most recent call last):
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\demo_gradio.py", line 126, in worker
    llama_vec, clip_l_pooler = encode_prompt_conds(prompt, text_encoder, text_encoder_2, tokenizer, tokenizer_2)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "G:\1\FramePack\framepack_cu126_torch26\framepack_cu126_torch26\webui\diffusers_helper\hunyuan.py", line 31, in encode_prompt_conds
    llama_attention_length = int(llama_attention_mask.sum())
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.

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

hahaha253

跪求各位老哥指导

970977470

啥都没有， 用的整合包是 谁发布的，支持 50系列的显卡版本吗

dtfour09

不太清楚，我也是5070ti跑的。但是只有跑眼罩有点问题。

hahaha253

970977470 发表于 2025-6-2 14:05
啥都没有， 用的整合包是 谁发布的，支持 50系列的显卡版本吗

大哥，就是滚石发的那个https://dfldata.cc/forum.php?mod ... &extra=page%3D1

hahaha253

970977470 发表于 2025-6-2 14:05
啥都没有， 用的整合包是 谁发布的，支持 50系列的显卡版本吗

就是滚石发布的那个

hahaha253

dtfour09 发表于 2025-6-2 15:19
不太清楚，我也是5070ti跑的。但是只有跑眼罩有点问题。

你也是用的滚石那个吗

hahaha253

970977470 发表于 2025-6-2 14:05
啥都没有， 用的整合包是 谁发布的，支持 50系列的显卡版本吗

问了一下DS

这个错误的核心原因是 GPU 计算能力不兼容，具体分析如下：

根本原因
GPU 架构不兼容：

log
NVIDIA GeForce RTX 5070 Ti with CUDA capability sm_120 is not compatible
The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90
您的 RTX 5070 Ti 需要 sm_120 架构支持

当前 PyTorch 版本最高只支持到 sm_90（对应 RTX 40 系列）

关键错误点：

log
RuntimeError: CUDA error: no kernel image is available for execution on the device
int(llama_attention_mask.sum())  # 触发CUDA内核调用
当尝试执行 attention_mask.sum() 时触发兼容性错误

这是模型计算注意力机制的核心操作

hahaha253

970977470 发表于 2025-6-2 14:05
啥都没有， 用的整合包是 谁发布的，支持 50系列的显卡版本吗

他的意思是PyTorch 版本版本太低了吗？要更新到最新版本吗？

hahaha253

dtfour09 发表于 2025-6-2 15:19
不太清楚，我也是5070ti跑的。但是只有跑眼罩有点问题。

他们说是没加模型，但是我下载的时候就是这个样子。没显示在哪里增加模型