#7229 翻译字幕阶段出错 [M2M100(本地内置)] parallel_for failed: cudaErrorNoKernelImageForDevice: no kernel image is available for execution

60.220**0 Posted at: 1 hour ago

翻译字幕阶段出错 [M2M100(本地内置)] parallel_for failed: cudaErrorNoKernelImageForDevice: no kernel image is available for execution on the device
Traceback (most recent call last):

File "videotrans\task\job.py", line 35, in run

File "videotrans\task\job.py", line 135, in process_task

File "videotrans\task\translate_srt.py", line 47, in trans

File "videotrans\translator\_runner.py", line 63, in run

File "videotrans\translator\_base.py", line 68, in run

File "videotrans\translator\_base.py", line 107, in _run_text

File "videotrans\translator\_m2m100.py", line 90, in _item_task

RuntimeError: parallel_for failed: cudaErrorNoKernelImageForDevice: no kernel image is available for execution on the device
cfg=TaskCfgSTS(uuid='bd443739a9', name='E:/AE/After Effects Guru/01 After Effects CC Essential Training(核心培训)/122431_00_02_3D15_files.srt', dirname='E:/AE/After Effects Guru/01 After Effects CC Essential Training(核心培训)', noextname='122431_00_02_3D15_files', basename='122431_00_02_3D15_files.srt', ext='srt', target_dir='E:/AE/After Effects Guru/01 After Effects CC Essential Training(核心培训)', cache_folder=None, is_cuda=False, source_language=None, source_language_code='en', source_sub='E:/AE/After Effects Guru/01 After Effects CC Essential Training(核心培训)/122431_00_02_3D15_files.srt', source_wav=None, source_wav_output=None, target_language=None, target_language_code='zh', target_sub='E:/AE/After Effects Guru/01 After Effects CC Essential Training(核心培训)/122431_00_02_3D15_files.zh.srt', target_wav=None, target_wav_output=None, translate_type=2)


=
system:Windows-10-10.0.19045-SP0
version:v4.06
frozen:True
language:zh
root_dir:D:/win-pyvideotrans-v4.06-0720

1 Replies

#1 DeepSeek reply at 1 hour ago

该错误的底层原因是:M2M100 本地翻译模型尝试使用 GPU(CUDA)运行,但当前显卡的算力架构与该模型预编译的 CUDA 内核不兼容,导致 cudaErrorNoKernelImageForDevice(设备上没有可执行的 kernel 镜像)。

M2M100 模型只要检测到 CUDA 可用,无论是否勾选“启用 CUDA 加速”,都会优先尝试使用 GPU,因此会绕过开关直接报错。

解决方案(任选其一):

  1. 改用其他翻译渠道(推荐):如 DeepSeek、Google(免费) 等,避免使用 M2M100 本地模型。
  2. 强制 M2M100 使用 CPU 运行:进入 菜单 → 工具/选项 → 高级选项 → 通用设置,将 强制指定重型任务运行设备 设置为 CPU,保存后重试。
  3. 升级显卡驱动:前往 NVIDIA 官网将显卡驱动更新到最新版,某些情况下可解决内核镜像不兼容问题。但若显卡过于老旧,仍可能无法兼容,建议直接采用方案 1 或 2。

请查阅相关文档:

Thinking...
Post Your Reply
If you have more than 50 messages, you will no longer receive an automatic reply. Please ask a new question!
Open source and free maintenance is not easy. If this project is helpful to you, please consider making a small donation to help the project continue to maintain and update.

Dondate via ko-fi.com

Related resource