MiniMax-H3 V3版：让视频创作变得超简单，提速50%，文字图片一键生成带声高清视频

无言***

xu87032040 发表于 2026-8-9 22:05
Press any key to continue . . .是啥原因啊

贴出完整的报错信息

z2***

轩辕 发表于 2026-8-7 21:16
大佬，这个版本支持NFSW吗？

想啥呢~

xu870***

无言以对 发表于 2026-8-10 09:15
贴出完整的报错信息

=====================================================
更多前沿AI技术和应用，访问 https://deepface.cc
软件WebUI正在启动中，请稍后...
ComfyUI服务器: http://127.0.0.1:8188
Gradio界面: http://localhost:7860
=====================================================

======================================================================
[11:06:34] [Launcher] MiniMax-H3 图生视频 — 启动器
[11:06:34] [Launcher] ComfyUI 端口: 8188
[11:06:34] [Launcher] Gradio  端口: 7860
[11:06:34] [Launcher] Python 路径:  D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\python.exe
======================================================================
[11:06:34] [Launcher] 启动 ComfyUI: D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\python.exe -u -s D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\main.py --listen 127.0.0.1 --port 8188
[11:06:34] [Launcher] 等待 ComfyUI 就绪...
[11:06:35] [ComfyUI] [INFO]
[11:06:35] [ComfyUI] Prestartup times for custom nodes:
[11:06:35] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-Easy-Use
[11:06:35] [ComfyUI] [INFO]
[11:06:37] [ComfyUI] [WARNING] WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations.
[11:06:37] [ComfyUI] [INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_embedding', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'int8_linear', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[11:06:37] [ComfyUI] [INFO] Found comfy_kitchen backend cuda: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'gemv_awq_w4a16', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8']}
[11:06:37] [ComfyUI] [INFO] Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'int8_linear', 'quantize_and_rotate_rowwise', 'quantize_int8_rowwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_']}
[11:06:37] [ComfyUI] [INFO] Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
[11:06:37] [ComfyUI] [INFO] Checkpoint files will always be loaded safely.
[11:06:37] [ComfyUI] [INFO] Total VRAM 24576 MB, total RAM 130864 MB
[11:06:37] [ComfyUI] [INFO] pytorch version: 2.7.1+cu128
[11:06:37] [ComfyUI] [INFO] Set vram state to: NORMAL_VRAM
[11:06:37] [ComfyUI] [INFO] Device: cuda:0 NVIDIA GeForce RTX 3090 : cudaMallocAsync
[11:06:37] [ComfyUI] [INFO] Using async weight offloading with 2 streams
[11:06:37] [ComfyUI] [INFO] Enabled pinned memory 52345.0
[11:06:38] [ComfyUI] [INFO] Using pytorch attention
[11:06:40] [ComfyUI] [WARNING] Unsupported Pytorch detected. DynamicVRAM support requires Pytorch version 2.8 or later. Falling back to legacy ModelPatcher. VRAM estimates may be unreliable especially on Windows
[11:06:40] [ComfyUI] [INFO] Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr  8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
[11:06:40] [ComfyUI] [INFO] ComfyUI version: 0.30.0
[11:06:40] [ComfyUI] [INFO] comfy-aimdo version: 0.4.13
[11:06:40] [ComfyUI] [INFO] comfy-kitchen version: 0.2.26
[11:06:40] [ComfyUI] [INFO] comfyui-frontend-package version: 1.48.6
[11:06:40] [ComfyUI] [INFO] comfyui-workflow-templates version: 0.11.31
[11:06:40] [ComfyUI] [INFO] comfyui-embedded-docs version: 0.5.9
[11:06:40] [ComfyUI] [INFO] comfy-kitchen version: 0.2.26
[11:06:40] [ComfyUI] [INFO] comfy-aimdo version: 0.4.13
[11:06:40] [ComfyUI] [INFO] [Prompt Server] web root: D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\Lib\site-packages\comfyui_frontend_package\static
[11:06:40] [ComfyUI] [INFO] Asset seeder disabled
[11:06:41] [ComfyUI] [INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
[11:06:43] [ComfyUI] [INFO] ComfyUI-GGUF: Partial torch compile only, consider updating pytorch
[11:06:43] [ComfyUI] [INFO]
[11:06:43] [ComfyUI] Import times for custom nodes:
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\websocket_image_save.py
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-MiniMax-H3-Turbo
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-GGUF
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-Spectrum-MiniMax-H3
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-SolAttn_triton
[11:06:43] [ComfyUI] [INFO]    0.0 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-KJNodes
[11:06:43] [ComfyUI] [INFO]    0.1 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\Nvidia_RTX_Nodes_ComfyUI
[11:06:43] [ComfyUI] [INFO]    0.1 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite
[11:06:43] [ComfyUI] [INFO]    1.4 seconds: D:\LeonAI\ComfyUI_MiniMax-H3-V3\ComfyUI\custom_nodes\ComfyUI-Easy-Use
[11:06:43] [ComfyUI] [INFO]
[11:06:43] [ComfyUI] [INFO] Context impl SQLiteImpl.
[11:06:43] [ComfyUI] [INFO] Will assume non-transactional DDL.
[11:06:43] [ComfyUI] [INFO] Using RAM pressure cache.
[11:06:43] [ComfyUI] [INFO] Starting server
[11:06:43] [ComfyUI] [INFO] To see the GUI go to: http://127.0.0.1:8188
[11:06:44] [Launcher] ComfyUI 已就绪!
[11:06:46] [Launcher] 启动 Gradio: D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\python.exe -u D:\LeonAI\ComfyUI_MiniMax-H3-V3\app.py
[11:06:48] [Gradio] ComfyUI 连接成功
[11:06:48] [Gradio] Traceback (most recent call last):
[11:06:48] [Gradio] File "<frozen __main__>", line 3, in <module>
[11:06:48] [Gradio] File "<frozen app>", line 933, in <module>
[11:06:48] [Gradio] File "D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\Lib\site-packages\gradio\blocks.py", line 2635, in launch
[11:06:48] [Gradio] ) = http_server.start_server(
[11:06:48] [Gradio] ^^^^^^^^^^^^^^^^^^^^^^^^^
[11:06:48] [Gradio] File "D:\LeonAI\ComfyUI_MiniMax-H3-V3\deepface\Lib\site-packages\gradio\http_server.py", line 157, in start_server
[11:06:48] [Gradio] raise OSError(
[11:06:48] [Gradio] OSError: Cannot find empty port in range: 7860-7860. You can specify a different port by setting the GRADIO_SERVER_PORT environment variable or passing the `server_port` parameter to `launch()`.
======================================================================
[11:06:51] [Launcher] 所有服务已启动!
[11:06:51] [Launcher] ComfyUI:  http://127.0.0.1:8188
[11:06:51] [Launcher] Gradio:   http://localhost:7860
======================================================================
[11:06:51] [Launcher] 按 Ctrl+C 停止所有服务
[11:06:51] [Launcher] 进程 143904 已退出 (返回码: 1)
[11:06:51] [Launcher] 正在关闭所有服务...
[11:06:51] [Launcher] 所有服务已关闭
[11:06:51] [Launcher] 正在关闭所有服务...
[11:06:51] [Launcher] 所有服务已关闭
Press any key to continue . . .

无言***

xu87032040 发表于 2026-8-10 11:07
=====================================================
更多前沿AI技术和应用，访问 https://deepface ...

端口被占用，其他的类似占用7860端口的程序退出再启动

轩***

xu87032040 发表于 2026-8-9 22:05
Press any key to continue . . .是啥原因啊

你后台还有一控制台没关

xu870***

无言以对 发表于 2026-8-10 11:19
端口被占用，其他的类似占用7860端口的程序退出再启动

原来如此，感谢

z2***

MiniMax-H3-FL2VA-Q4_K_M.gguf这个模型没有~

无言***

z2625 发表于 2026-8-10 17:14
MiniMax-H3-FL2VA-Q4_K_M.gguf这个模型没有~

这个包根本用不到这个模型。
主要用的是参考模型

wjjd***

进程 33980 已退出 (返回码: 1)，每次都出现一大片蓝色代码，然后就停了

无言***

wjjd1991 发表于 2026-8-10 19:21
进程 33980 已退出 (返回码: 1)，每次都出现一大片蓝色代码，然后就停了

先看显卡是否符合要求
如果是新手，先去看看新手必看，确认不是以上问题
贴出完整报错代码