1 line
106 KiB
JSON
1 line
106 KiB
JSON
{"69447201-c928-40a2-8f9b-183427af45b2": {"prompt": [144, "69447201-c928-40a2-8f9b-183427af45b2", {"1": {"class_type": "UNETLoader", "inputs": {"unet_name": "minimax_h3_fl2va_int8_convrot.safetensors", "weight_dtype": "default"}, "_meta": {"title": "MiniMax H3 FL2VA INT8 \u6a21\u578b"}}, "2": {"class_type": "CLIPLoader", "inputs": {"clip_name": "qwen3vl_32b_minimax_h3_int8_convrot.safetensors", "type": "minimax", "device": "default"}, "_meta": {"title": "Qwen3-VL MiniMax H3 INT8 \u6587\u672c\u7f16\u7801\u5668"}}, "3": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_video_vae_fp16.safetensors"}, "_meta": {"title": "MiniMax H3 \u89c6\u9891 VAE"}}, "4": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_audio_vae_fp32.safetensors"}, "_meta": {"title": "MiniMax H3 \u97f3\u9891 VAE"}}, "5": {"class_type": "MiniMaxH3ReferenceToVideo", "inputs": {"clip": ["2", 0], "vae": ["3", 0], "audio_vae": ["4", 0], "prompt": "photorealistic live-action film still, Chinese actors with authentic East Asian facial features, natural facial features and realistic human proportions, authentic skin texture with visible pores and subtle imperfections, natural yellowish skin tones, lifelike East Asian eyes and individual hair strands, realistic fabric weave and material detail, grounded real-world locations with believable scale and spatial depth, professional cinematic photography, physically plausible lighting motivated by the scene, soft natural shadow transitions, controlled highlights and rich tonal detail, restrained cinematic color grading, subtle film grain, realistic lens perspective and depth of field appropriate to the requested framing, natural expressions and body language, consistent actor identity and appearance across shots, preserve the character's specified age, costume and story period, avoid anime, cartoons, illustration, 3D animation, game-engine rendering, doll-like faces, plastic or waxy skin, excessive beauty retouching and exaggerated anatomy\uff0c\n0-3s:Frontal eye-level slightly low-angle wide establishing empty shot, early-morning ancient-style courtyard, a flat bluestone-paved small courtyard at the center of frame, fine green moss growing in the stone crevices, a peach blossom tree slanting out from the right side of frame over a plain white-pink wall, pink-white peach blossoms blown down by the breeze, petals drifting onto the bluestone slabs and to the edge of the grey-blue water vat in the courtyard corner, thin morning mist floating over the central open area, cool white pale-gold skylight slanting down from the upper side, mottled peach tree shadows cast on the ground, throughout only natural ambient sound effects of early-morning breeze, birdsong and petals lightly falling are kept.\n3-5s:The wide shot slowly pushes in, wind passes the peach branches, a rain of petals falls, spreading a thin layer of pink-white over the bluestone slabs, fine ripples spread across the water surface of the water vat in the courtyard corner, morning mist drifts gently in the cool white pale-gold light, peach tree shadows sway gently with the breeze, the bluestone open area is flat and open, throughout only natural ambient sound effects are kept.\n[Continuity: visual guidance only; these notes are not spoken.]\nStart: Opening: early-morning ancient-style courtyard, frontal eye-level slightly low-angle wide empty shot; in the center of the bluestone-paved small yard floats thin morning mist; a peach tree leans out from the right over a plain white pinkish wall; pink-white petals fall; cool white pale-gold skylight slants in; a blue-gray water vat stands still in the yard corner; the image is clear and quiet.\nEnd: The camera stops on the bluestone ground; petals spread in a thin pink-white layer; fine ripples spread across the water vat's surface; morning mist drifts gently; the peach tree's shadow sways slightly with the breeze; the bluestone open area is flat and broad.\n\n[HB_SCENE_LOCK_V1]\nVisual only; never spoken/displayed.\nSet: <Picture 1> \u53e4\u98ce\u5ead\u9662\nTime: early morning\nSetting: small courtyard paved with bluestone slabs; fine moss grows in the gaps; a peach tree leans out over the white wall, branches full of pink-white blossoms; a light breeze sends petals drifting down one by one, scattered on the slabs and beside a blue-gray water vat in the courtyard corner. Plain white plaster walls with dark gray tile eaves; a few low tufts of grass and moss marks at the wall base; overall a refined, understated ancient-style courtyard layout, open with ample negative space; a flat bluestone clearing in the center for dancing; clean, uncluttered background.\nLighting: soft early-morning natural daylight, cool white with a faint golden tint, slanting into the courtyard from above at the side; the peach tree casts faint dappled shadows on the bluestone slabs; overall bright and clear, soft light-to-dark transitions, a thin morning-mist feel in the air, petals semi-translucent in the light; sweet, gentle, quiet, healing atmosphere.\nImage 1: ENVIRONMENT only, never identity. Keep architecture, materials, furniture and object layout. Ignore character-sheet backgrounds.\nAll cuts/segments, close-up/reverse included: keep geometry, doors/windows, materials, furniture, prop/light positions; off-screen stays off-screen. Reframe within this set. Scripted prop/light changes persist. Empty sets never remove scripted actors/actions.\n[/HB_SCENE_LOCK_V1]\nReference image mapping (visual only): 1=environment <Picture 1> \u53e4\u98ce\u5ead\u9662.\nEnvironment images define architecture, not an empty cast or a shot to insert. Follow the scripted framing and preserve present actors until a motivated exit/cut; never replace them with the empty reference view.\n\nAudio direction: no speech. No dialogue. No narration. Only scripted ambience and nonverbal reactions; no invented words or subtitles.", "width": 1280, "height": 704, "length": ["7", 1], "ref_image_size": "max", "ref_images.ref_image_0": ["canvas_reference_1", 0]}, "_meta": {"title": "MiniMax H3 \u5168\u80fd\u591a\u53c2\u8003\u56fe\u751f\u89c6\u9891"}}, "6": {"class_type": "PrimitiveFloat", "inputs": {"value": 5.0}, "_meta": {"title": "\u65f6\u957f\uff08\u79d2\uff09"}}, "7": {"class_type": "ComfyMathExpression", "inputs": {"expression": "max(5, round(a * 24)) + (5 - (max(5, round(a * 24)) % 17)) % 17", "values.a": ["6", 0]}, "_meta": {"title": "H3 \u5e27\u6570\u5bf9\u9f50\uff0824fps\uff0c17k+5\uff09"}}, "8": {"class_type": "RandomNoise", "inputs": {"noise_seed": 7462808019464368227}, "_meta": {"title": "\u968f\u673a\u79cd\u5b50"}}, "9": {"class_type": "MiniMaxH3MemoryEfficientSageAttentionPatch", "inputs": {"model": ["1", 0]}, "_meta": {"title": "MiniMax H3 SageAttention \u663e\u5b58\u4f18\u5316"}}, "10": {"class_type": "LoraLoaderModelOnly", "inputs": {"model": ["9", 0], "lora_name": "minimax_h3_turbo_4\u6b65\u52a0\u901f_comfyui.safetensors", "strength_model": 1.0}, "_meta": {"title": "MiniMax H3 4\u6b65\u52a0\u901f LoRA"}}, "11": {"class_type": "BasicScheduler", "inputs": {"model": ["10", 0], "scheduler": "beta", "steps": 8, "denoise": 1.0}, "_meta": {"title": "MiniMax H3 Turbo \u8c03\u5ea6\u5668"}}, "12": {"class_type": "KSamplerSelect", "inputs": {"sampler_name": "euler"}, "_meta": {"title": "MiniMax H3 Turbo \u91c7\u6837\u5668"}}, "13": {"class_type": "BasicGuider", "inputs": {"model": ["10", 0], "conditioning": ["5", 0]}, "_meta": {"title": "MiniMax H3 \u5f15\u5bfc"}}, "14": {"class_type": "SamplerCustomAdvanced", "inputs": {"noise": ["8", 0], "guider": ["13", 0], "sampler": ["12", 0], "sigmas": ["11", 0], "latent_image": ["5", 1]}, "_meta": {"title": "MiniMax H3 Turbo \u9ad8\u7ea7\u91c7\u6837"}}, "15": {"class_type": "VAEDecode", "inputs": {"samples": ["14", 0], "vae": ["3", 0]}, "_meta": {"title": "\u89e3\u7801\u89c6\u9891"}}, "16": {"class_type": "VAEDecodeAudio", "inputs": {"samples": ["14", 0], "vae": ["4", 0]}, "_meta": {"title": "\u89e3\u7801\u539f\u751f\u97f3\u9891"}}, "17": {"class_type": "CreateVideo", "inputs": {"images": ["15", 0], "audio": ["16", 0], "fps": 24.0, "bit_depth": 8}, "_meta": {"title": "\u5408\u6210 H3 \u97f3\u89c6\u9891"}}, "18": {"class_type": "SaveVideo", "inputs": {"video": ["17", 0], "filename_prefix": "video/MiniMax_H3_AllReference_4step_canvas-1deaea39db8e5c31", "format": "auto", "codec": "auto"}, "_meta": {"title": "\u4fdd\u5b58 MiniMax H3 \u5168\u80fd\u53c2\u8003\u89c6\u9891"}}, "canvas_reference_1": {"class_type": "LoadImage", "inputs": {"image": "canvas-1deaea39db8e5c31-reference-1.png"}, "_meta": {"title": "Canvas reference 1"}}}, {"client_id": "canvas-1deaea39db8e5c31", "create_time": 1795085163330}, ["18"]], "outputs": {}, "status": {"status_str": "error", "completed": false, "messages": [["execution_start", {"prompt_id": "69447201-c928-40a2-8f9b-183427af45b2", "timestamp": 1795085163332}], ["execution_cached", {"nodes": ["1", "2", "3", "4", "9", "10", "11", "12"], "prompt_id": "69447201-c928-40a2-8f9b-183427af45b2", "timestamp": 1795085163347}], ["execution_error", {"prompt_id": "69447201-c928-40a2-8f9b-183427af45b2", "node_id": "14", "node_type": "SamplerCustomAdvanced", "executed": ["13", "8", "5", "canvas_reference_1", "7", "6"], "exception_message": "Cannot access data pointer of Tensor that doesn't have storage\nException raised from throw_data_ptr_access_error at /pytorch/c10/core/TensorImpl.cpp:307 (most recent call first):\nframe #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x9d (0x75ad7117305d in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, char const*) + 0x69 (0x75ad711077d3 in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #2: c10::TensorImpl::throw_data_ptr_access_error() const + 0x34 (0x75ad7114c634 in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #3: scale_fuse_quant_cuda(at::Tensor, at::Tensor, at::Tensor, int, float, int) + 0x1bb8 (0x75a9d12de358 in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #4: <unknown function> + 0x3bb69 (0x75a9d12f7b69 in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #5: <unknown function> + 0x38d57 (0x75a9d12f4d57 in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #6: <unknown function> + 0x1f9196 (0x6177194ad196 in /opt/conda/bin/python3)\nframe #7: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #8: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #9: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #10: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #11: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #12: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #13: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #14: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #15: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #16: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #17: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #18: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #19: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #20: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #21: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #22: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #23: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #24: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #25: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #26: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #27: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #28: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #29: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #30: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #31: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #32: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #33: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #34: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #35: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #36: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #37: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #38: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #39: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #40: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #41: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #42: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #43: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #44: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #45: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #46: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #47: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #48: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #49: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #50: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #51: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #52: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #53: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #54: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #55: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #56: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #57: PyObject_Call + 0xbe (0x6177194c68ee in /opt/conda/bin/python3)\nframe #58: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #59: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #60: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #61: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #62: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\n\n", "exception_type": "RuntimeError", "traceback": [" File \"/app/execution.py\", line 547, in execute\n output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 352, in get_output_data\n return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 326, in _async_map_node_over_list\n await process_inputs(input_dict, i)\n", " File \"/app/execution.py\", line 314, in process_inputs\n result = f(**inputs)\n ^^^^^^^^^^^\n", " File \"/app/comfy_api/internal/__init__.py\", line 149, in wrapped_func\n return method(locked_class, **inputs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_api/latest/_io.py\", line 2053, in EXECUTE_NORMALIZED\n to_return = cls.execute(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_extras/nodes_custom_sampler.py\", line 1055, in execute\n samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1335, in sample\n output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1262, in outer_sample\n output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1237, in inner_sample\n samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1005, in sample\n samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/utils/_contextlib.py\", line 124, in decorate_context\n return func(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/k_diffusion/sampling.py\", line 205, in sample_euler\n denoised = model(x, sigma_hat * s_in, **extra_args)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 640, in __call__\n out = self.inner_model(x, sigma, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1208, in __call__\n return self.outer_predict_noise(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1215, in outer_predict_noise\n ).execute(x, timestep, model_options, seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1218, in predict_noise\n return sampling_function(self.inner_model, x, timestep, self.conds.get(\"negative\", None), self.conds.get(\"positive\", None), self.cfg, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 620, in sampling_function\n out = calc_cond_batch(model, conds, x, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 211, in calc_cond_batch\n return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 219, in _calc_cond_batch_outer\n return executor.execute(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 335, in _calc_cond_batch\n output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 210, in apply_model\n return comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 254, in _apply_model\n model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 577, in forward\n graph_out = comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 764, in _forward\n h = block(h, t_emb, mod_segments, rope_freqs, transformer_options=transformer_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 295, in forward\n x = _mod_gate(x, gate_msa, attention(h, rope_freqs=rope_freqs, transformer_options=transformer_options), mod_segments)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/custom_nodes/comfyui-kjnodes/nodes/ltxv_nodes.py\", line 2106, in minimax_sageattn_forward\n o = _sageattn_int8_fp8_nhd(qkv, dtype)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/custom_nodes/comfyui-kjnodes/nodes/ltxv_nodes.py\", line 1947, in _sageattn_int8_fp8_nhd\n v_fp8, v_scale, _ = per_channel_fp8(v, tensor_layout=tensor_layout, scale_max=quant_v_scale_max, smooth_v=False)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/sageattention/quant.py\", line 292, in per_channel_fp8\n _fused.scale_fuse_quant_cuda(v_transposed_permutted, v_fp8, v_scale, kv_len, scale_max, _tensor_layout)\n"], "current_inputs": {"noise": ["</app/comfy_extras/nodes_custom_sampler.Noise_RandomNoise object at 0x75a4c45bc5d0>"], "guider": ["</app/comfy_extras/nodes_custom_sampler.Guider_Basic object at 0x75a27074c150>"], "sampler": ["<comfy.samplers.KSAMPLER object at 0x75a48da3de10>"], "sigmas": ["tensor([1.0000, 0.9950, 0.9825, 0.9607, 0.9234, 0.8553, 0.7207, 0.4249, 0.0000])"], "latent_image": ["{'samples': NestedTensor([tensor([[[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n ...,\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]]]), tensor([[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]])])}"]}, "current_outputs": ["8", "2", "3", "10", "1", "4", "13", "5", "11", "17", "6", "18", "14", "9", "canvas_reference_1", "12", "16", "7", "15"], "timestamp": 1795085170434}]]}, "meta": {}}, "15773e33-c4ff-4ec3-a673-9dbda6075818": {"prompt": [145, "15773e33-c4ff-4ec3-a673-9dbda6075818", {"1": {"class_type": "UNETLoader", "inputs": {"unet_name": "minimax_h3_fl2va_int8_convrot.safetensors", "weight_dtype": "default"}, "_meta": {"title": "MiniMax H3 FL2VA INT8 \u6a21\u578b"}}, "2": {"class_type": "CLIPLoader", "inputs": {"clip_name": "qwen3vl_32b_minimax_h3_int8_convrot.safetensors", "type": "minimax", "device": "default"}, "_meta": {"title": "Qwen3-VL MiniMax H3 INT8 \u6587\u672c\u7f16\u7801\u5668"}}, "3": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_video_vae_fp16.safetensors"}, "_meta": {"title": "MiniMax H3 \u89c6\u9891 VAE"}}, "4": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_audio_vae_fp32.safetensors"}, "_meta": {"title": "MiniMax H3 \u97f3\u9891 VAE"}}, "5": {"class_type": "MiniMaxH3ReferenceToVideo", "inputs": {"clip": ["2", 0], "vae": ["3", 0], "audio_vae": ["4", 0], "prompt": "photorealistic live-action film still, Chinese actors with authentic East Asian facial features, natural facial features and realistic human proportions, authentic skin texture with visible pores and subtle imperfections, natural yellowish skin tones, lifelike East Asian eyes and individual hair strands, realistic fabric weave and material detail, grounded real-world locations with believable scale and spatial depth, professional cinematic photography, physically plausible lighting motivated by the scene, soft natural shadow transitions, controlled highlights and rich tonal detail, restrained cinematic color grading, subtle film grain, realistic lens perspective and depth of field appropriate to the requested framing, natural expressions and body language, consistent actor identity and appearance across shots, preserve the character's specified age, costume and story period, avoid anime, cartoons, illustration, 3D animation, game-engine rendering, doll-like faces, plastic or waxy skin, excessive beauty retouching and exaggerated anatomy\uff0c\n0-3s:Frontal eye-level slightly low-angle wide establishing empty shot, early-morning ancient-style courtyard, a flat bluestone-paved small courtyard at the center of frame, fine green moss growing in the stone crevices, a peach blossom tree slanting out from the right side of frame over a plain white pinkish wall, pink-white peach blossoms blown down by the breeze, petals drifting onto the stone slabs and to the edge of the grey-blue water vat in the courtyard corner, thin morning mist floating over the central open area, cool white pale-gold skylight slanting down from the upper side, dappled peach tree shadows cast on the ground, throughout only ambient natural sound effects of early-morning breeze, birdsong and petals lightly falling are kept.\n3-5s:The wide shot slowly pushes in, wind passes the peach branches, a rain of petals falls, spreading a thin layer of pink-white over the bluestone slabs, fine ripples spread across the water surface of the courtyard-corner vat, morning mist drifts gently in the cool white pale-gold light, peach tree shadows sway gently with the breeze, the bluestone open area is flat and broad, throughout only ambient natural sound effects are kept. No text appears anywhere in the frame.\n[Continuity: visual guidance only; these notes are not spoken.]\nStart: Opening: early-morning ancient-style courtyard, frontal eye-level slightly low-angle wide establishing empty shot; in center a bluestone-paved small yard with thin morning mist; a peach tree slanting out from the right over a plain white-powdered wall; pink-white petals falling; cool white pale-gold skylight slanting in; a blue-gray water vat standing still in the yard corner; image clear and quiet.\nEnd: Camera stops on the bluestone ground; petals spread in a thin pink-white layer; fine ripples spread across the vat's water surface; morning mist drifts gently; peach-tree shadows sway slightly in the breeze; the bluestone open area is flat and broad.\n\n[HB_SCENE_LOCK_V1]\nVisual only; never spoken/displayed.\nSet: <Picture 1> \u53e4\u98ce\u5ead\u9662\nTime: early morning\nSetting: small courtyard paved with bluestone slabs; fine moss grows in the gaps; a peach tree leans out over the white wall, branches full of pink-white blossoms; a light breeze sends petals drifting down one by one, scattered on the slabs and beside a blue-gray water vat in the courtyard corner. Plain white plaster walls with dark gray tile eaves; a few low tufts of grass and moss marks at the wall base; overall a refined, understated ancient-style courtyard layout, open with ample negative space; a flat bluestone clearing in the center for dancing; clean, uncluttered background.\nLighting: soft early-morning natural daylight, cool white with a faint golden tint, slanting into the courtyard from above at the side; the peach tree casts faint dappled shadows on the bluestone slabs; overall bright and clear, soft light-to-dark transitions, a thin morning-mist feel in the air, petals semi-translucent in the light; sweet, gentle, quiet, healing atmosphere.\nImage 1: ENVIRONMENT only, never identity. Keep architecture, materials, furniture and object layout. Ignore character-sheet backgrounds.\nAll cuts/segments, close-up/reverse included: keep geometry, doors/windows, materials, furniture, prop/light positions; off-screen stays off-screen. Reframe within this set. Scripted prop/light changes persist. Empty sets never remove scripted actors/actions.\n[/HB_SCENE_LOCK_V1]\nReference image mapping (visual only): 1=environment <Picture 1> \u53e4\u98ce\u5ead\u9662.\nEnvironment images define architecture, not an empty cast or a shot to insert. Follow the scripted framing and preserve present actors until a motivated exit/cut; never replace them with the empty reference view.\n\nAudio direction: no speech. No dialogue. No narration. Only scripted ambience and nonverbal reactions; no invented words or subtitles.", "width": 1280, "height": 704, "length": ["7", 1], "ref_image_size": "max", "ref_images.ref_image_0": ["canvas_reference_1", 0]}, "_meta": {"title": "MiniMax H3 \u5168\u80fd\u591a\u53c2\u8003\u56fe\u751f\u89c6\u9891"}}, "6": {"class_type": "PrimitiveFloat", "inputs": {"value": 5.0}, "_meta": {"title": "\u65f6\u957f\uff08\u79d2\uff09"}}, "7": {"class_type": "ComfyMathExpression", "inputs": {"expression": "max(5, round(a * 24)) + (5 - (max(5, round(a * 24)) % 17)) % 17", "values.a": ["6", 0]}, "_meta": {"title": "H3 \u5e27\u6570\u5bf9\u9f50\uff0824fps\uff0c17k+5\uff09"}}, "8": {"class_type": "RandomNoise", "inputs": {"noise_seed": 573220205686573698}, "_meta": {"title": "\u968f\u673a\u79cd\u5b50"}}, "9": {"class_type": "MiniMaxH3MemoryEfficientSageAttentionPatch", "inputs": {"model": ["1", 0]}, "_meta": {"title": "MiniMax H3 SageAttention \u663e\u5b58\u4f18\u5316"}}, "10": {"class_type": "LoraLoaderModelOnly", "inputs": {"model": ["9", 0], "lora_name": "minimax_h3_turbo_4\u6b65\u52a0\u901f_comfyui.safetensors", "strength_model": 1.0}, "_meta": {"title": "MiniMax H3 4\u6b65\u52a0\u901f LoRA"}}, "11": {"class_type": "BasicScheduler", "inputs": {"model": ["10", 0], "scheduler": "beta", "steps": 8, "denoise": 1.0}, "_meta": {"title": "MiniMax H3 Turbo \u8c03\u5ea6\u5668"}}, "12": {"class_type": "KSamplerSelect", "inputs": {"sampler_name": "euler"}, "_meta": {"title": "MiniMax H3 Turbo \u91c7\u6837\u5668"}}, "13": {"class_type": "BasicGuider", "inputs": {"model": ["10", 0], "conditioning": ["5", 0]}, "_meta": {"title": "MiniMax H3 \u5f15\u5bfc"}}, "14": {"class_type": "SamplerCustomAdvanced", "inputs": {"noise": ["8", 0], "guider": ["13", 0], "sampler": ["12", 0], "sigmas": ["11", 0], "latent_image": ["5", 1]}, "_meta": {"title": "MiniMax H3 Turbo \u9ad8\u7ea7\u91c7\u6837"}}, "15": {"class_type": "VAEDecode", "inputs": {"samples": ["14", 0], "vae": ["3", 0]}, "_meta": {"title": "\u89e3\u7801\u89c6\u9891"}}, "16": {"class_type": "VAEDecodeAudio", "inputs": {"samples": ["14", 0], "vae": ["4", 0]}, "_meta": {"title": "\u89e3\u7801\u539f\u751f\u97f3\u9891"}}, "17": {"class_type": "CreateVideo", "inputs": {"images": ["15", 0], "audio": ["16", 0], "fps": 24.0, "bit_depth": 8}, "_meta": {"title": "\u5408\u6210 H3 \u97f3\u89c6\u9891"}}, "18": {"class_type": "SaveVideo", "inputs": {"video": ["17", 0], "filename_prefix": "video/MiniMax_H3_AllReference_4step_canvas-bdda3f02fd52c6f0", "format": "auto", "codec": "auto"}, "_meta": {"title": "\u4fdd\u5b58 MiniMax H3 \u5168\u80fd\u53c2\u8003\u89c6\u9891"}}, "canvas_reference_1": {"class_type": "LoadImage", "inputs": {"image": "canvas-bdda3f02fd52c6f0-reference-1.png"}, "_meta": {"title": "Canvas reference 1"}}}, {"client_id": "canvas-bdda3f02fd52c6f0", "create_time": 1795085181805}, ["18"]], "outputs": {}, "status": {"status_str": "error", "completed": false, "messages": [["execution_start", {"prompt_id": "15773e33-c4ff-4ec3-a673-9dbda6075818", "timestamp": 1795085181807}], ["execution_cached", {"nodes": ["1", "2", "3", "4", "6", "7", "9", "10", "11", "12"], "prompt_id": "15773e33-c4ff-4ec3-a673-9dbda6075818", "timestamp": 1795085181822}], ["execution_error", {"prompt_id": "15773e33-c4ff-4ec3-a673-9dbda6075818", "node_id": "14", "node_type": "SamplerCustomAdvanced", "executed": ["5", "8", "13", "canvas_reference_1"], "exception_message": "Cannot access data pointer of Tensor that doesn't have storage\nException raised from throw_data_ptr_access_error at /pytorch/c10/core/TensorImpl.cpp:307 (most recent call first):\nframe #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x9d (0x75ad7117305d in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, char const*) + 0x69 (0x75ad711077d3 in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #2: c10::TensorImpl::throw_data_ptr_access_error() const + 0x34 (0x75ad7114c634 in /opt/conda/lib/python3.11/site-packages/torch/lib/libc10.so)\nframe #3: void* c10::TensorImpl::data_impl<void, c10::TensorImpl::mutable_data()::{lambda()#1}>(c10::TensorImpl::mutable_data()::{lambda()#1} const&) const + 0xa8 (0x75a9d12e6ae8 in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #4: quant_per_block_int8_fuse_sub_mean_cuda(at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int) + 0x1c0f (0x75a9d12d348f in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #5: <unknown function> + 0x3c55c (0x75a9d12f855c in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #6: <unknown function> + 0x38d57 (0x75a9d12f4d57 in /opt/conda/lib/python3.11/site-packages/sageattention/_fused.cpython-311-x86_64-linux-gnu.so)\nframe #7: <unknown function> + 0x1f9196 (0x6177194ad196 in /opt/conda/bin/python3)\nframe #8: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #9: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #10: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #11: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #12: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #13: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #14: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #15: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #16: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #17: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #18: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #19: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #20: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #21: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #22: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #23: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #24: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #25: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #26: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #27: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #28: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #29: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #30: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #31: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #32: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #33: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #34: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #35: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #36: _PyObject_MakeTpCall + 0x29b (0x61771948a6ab in /opt/conda/bin/python3)\nframe #37: _PyEval_EvalFrameDefault + 0x70a (0x617719497d2a in /opt/conda/bin/python3)\nframe #38: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #39: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #40: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #41: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #42: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #43: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #44: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #45: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #46: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #47: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #48: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #49: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #50: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #51: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #52: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\nframe #53: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #54: _PyFunction_Vectorcall + 0x17f (0x6177194bca4f in /opt/conda/bin/python3)\nframe #55: _PyObject_FastCallDictTstate + 0xe0 (0x61771948f320 in /opt/conda/bin/python3)\nframe #56: _PyObject_Call_Prepend + 0x69 (0x6177194c44c9 in /opt/conda/bin/python3)\nframe #57: <unknown function> + 0x2df799 (0x617719593799 in /opt/conda/bin/python3)\nframe #58: PyObject_Call + 0xbe (0x6177194c68ee in /opt/conda/bin/python3)\nframe #59: _PyEval_EvalFrameDefault + 0x4a29 (0x61771949c049 in /opt/conda/bin/python3)\nframe #60: <unknown function> + 0x227a92 (0x6177194dba92 in /opt/conda/bin/python3)\nframe #61: <unknown function> + 0x227316 (0x6177194db316 in /opt/conda/bin/python3)\nframe #62: PyObject_Call + 0x134 (0x6177194c6964 in /opt/conda/bin/python3)\n\n", "exception_type": "RuntimeError", "traceback": [" File \"/app/execution.py\", line 547, in execute\n output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 352, in get_output_data\n return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 326, in _async_map_node_over_list\n await process_inputs(input_dict, i)\n", " File \"/app/execution.py\", line 314, in process_inputs\n result = f(**inputs)\n ^^^^^^^^^^^\n", " File \"/app/comfy_api/internal/__init__.py\", line 149, in wrapped_func\n return method(locked_class, **inputs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_api/latest/_io.py\", line 2053, in EXECUTE_NORMALIZED\n to_return = cls.execute(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_extras/nodes_custom_sampler.py\", line 1055, in execute\n samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1335, in sample\n output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1262, in outer_sample\n output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1237, in inner_sample\n samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1005, in sample\n samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/utils/_contextlib.py\", line 124, in decorate_context\n return func(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/k_diffusion/sampling.py\", line 205, in sample_euler\n denoised = model(x, sigma_hat * s_in, **extra_args)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 640, in __call__\n out = self.inner_model(x, sigma, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1208, in __call__\n return self.outer_predict_noise(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1215, in outer_predict_noise\n ).execute(x, timestep, model_options, seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1218, in predict_noise\n return sampling_function(self.inner_model, x, timestep, self.conds.get(\"negative\", None), self.conds.get(\"positive\", None), self.cfg, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 620, in sampling_function\n out = calc_cond_batch(model, conds, x, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 211, in calc_cond_batch\n return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 219, in _calc_cond_batch_outer\n return executor.execute(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 335, in _calc_cond_batch\n output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 210, in apply_model\n return comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 254, in _apply_model\n model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 577, in forward\n graph_out = comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 764, in _forward\n h = block(h, t_emb, mod_segments, rope_freqs, transformer_options=transformer_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 295, in forward\n x = _mod_gate(x, gate_msa, attention(h, rope_freqs=rope_freqs, transformer_options=transformer_options), mod_segments)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/custom_nodes/comfyui-kjnodes/nodes/ltxv_nodes.py\", line 2106, in minimax_sageattn_forward\n o = _sageattn_int8_fp8_nhd(qkv, dtype)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/custom_nodes/comfyui-kjnodes/nodes/ltxv_nodes.py\", line 1945, in _sageattn_int8_fp8_nhd\n q_int8, q_scale, k_int8, k_scale = per_warp_int8_cuda(q, k, km=k.mean(dim=1, keepdim=True), tensor_layout=tensor_layout, BLKQ=128, WARPQ=32, BLKK=64)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/sageattention/quant.py\", line 176, in per_warp_int8\n _fused.quant_per_block_int8_fuse_sub_mean_cuda(k, km, k_int8, k_scale, BLKK, _tensor_layout)\n"], "current_inputs": {"noise": ["</app/comfy_extras/nodes_custom_sampler.Noise_RandomNoise object at 0x75a48c479f90>"], "guider": ["</app/comfy_extras/nodes_custom_sampler.Guider_Basic object at 0x75a4381c6150>"], "sampler": ["<comfy.samplers.KSAMPLER object at 0x75a48da3de10>"], "sigmas": ["tensor([1.0000, 0.9950, 0.9825, 0.9607, 0.9234, 0.8553, 0.7207, 0.4249, 0.0000])"], "latent_image": ["{'samples': NestedTensor([tensor([[[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n ...,\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]]]), tensor([[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]])])}"]}, "current_outputs": ["8", "2", "3", "10", "1", "4", "13", "5", "11", "17", "6", "18", "14", "9", "canvas_reference_1", "12", "16", "7", "15"], "timestamp": 1795085187795}]]}, "meta": {}}, "a40ef771-5ec1-4b65-816d-41ec318e60aa": {"prompt": [146, "a40ef771-5ec1-4b65-816d-41ec318e60aa", {"1": {"class_type": "UNETLoader", "inputs": {"unet_name": "minimax_h3_fl2va_int8_convrot.safetensors", "weight_dtype": "default"}, "_meta": {"title": "MiniMax H3 FL2VA INT8 \u6a21\u578b"}}, "2": {"class_type": "CLIPLoader", "inputs": {"clip_name": "qwen3vl_32b_minimax_h3_int8_convrot.safetensors", "type": "minimax", "device": "default"}, "_meta": {"title": "Qwen3-VL MiniMax H3 INT8 \u6587\u672c\u7f16\u7801\u5668"}}, "3": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_video_vae_fp16.safetensors"}, "_meta": {"title": "MiniMax H3 \u89c6\u9891 VAE"}}, "4": {"class_type": "VAELoader", "inputs": {"vae_name": "minimax_h3_audio_vae_fp32.safetensors"}, "_meta": {"title": "MiniMax H3 \u97f3\u9891 VAE"}}, "5": {"class_type": "MiniMaxH3ReferenceToVideo", "inputs": {"clip": ["2", 0], "vae": ["3", 0], "audio_vae": ["4", 0], "prompt": "photorealistic live-action film still, Chinese actors with authentic East Asian facial features, natural facial features and realistic human proportions, authentic skin texture with visible pores and subtle imperfections, natural yellowish skin tones, lifelike East Asian eyes and individual hair strands, realistic fabric weave and material detail, grounded real-world locations with believable scale and spatial depth, professional cinematic photography, physically plausible lighting motivated by the scene, soft natural shadow transitions, controlled highlights and rich tonal detail, restrained cinematic color grading, subtle film grain, realistic lens perspective and depth of field appropriate to the requested framing, natural expressions and body language, consistent actor identity and appearance across shots, preserve the character's specified age, costume and story period, avoid anime, cartoons, illustration, 3D animation, game-engine rendering, doll-like faces, plastic or waxy skin, excessive beauty retouching and exaggerated anatomy\uff0c\n0-3s:Frontal eye-level slightly low-angle wide establishing empty shot, early-morning ancient-style courtyard, a flat bluestone-paved small courtyard at the center of the frame, fine green moss growing between the stone seams, a peach tree slanting out from the right side of the frame over a plain white pinkish wall, pink-white peach blossoms blown down by the breeze, petals drifting one by one onto the stone slabs and over to the gray-blue water vat in the courtyard corner, thin morning mist floating over the central open area, cold white pale-gold skylight slanting down from the upper side, mottled peach-tree shadows cast on the ground.\n3-5s:Wide shot slowly pushing in, wind passing through the peach branches, a shower of petals falls, spreading a thin layer of pink-white over the bluestone slabs, fine ripples spreading across the water surface of the courtyard-corner vat, morning mist drifting gently in the cold white pale-gold light, peach-tree shadows swaying gently with the breeze, the bluestone open area flat and broad. No text of any kind appears in the frame.\n[Continuity: visual guidance only; these notes are not spoken.]\nStart: Opening: early morning, ancient-style courtyard, frontal eye-level slightly low-angle wide establishing empty shot; in center a bluestone-paved small courtyard with thin morning mist; a peach tree slanting out from the right over a plain white pink wall; pink-white petals falling; cool white pale-gold skylight slanting in; a blue-gray water vat standing still in the courtyard corner; image clear and quiet.\nEnd: Camera stops on the bluestone ground, petals spread in a thin pink-white layer, fine ripples spread on the water vat's surface, morning mist drifts gently, peach tree shadows sway slightly in the breeze, the bluestone open area flat and broad.\n\n[HB_SCENE_LOCK_V1]\nVisual only; never spoken/displayed.\nSet: <Picture 1> \u53e4\u98ce\u5ead\u9662\nTime: early morning\nSetting: small courtyard paved with bluestone slabs; fine moss grows in the gaps; a peach tree leans out over the white wall, branches full of pink-white blossoms; a light breeze sends petals drifting down one by one, scattered on the slabs and beside a blue-gray water vat in the courtyard corner. Plain white plaster walls with dark gray tile eaves; a few low tufts of grass and moss marks at the wall base; overall a refined, understated ancient-style courtyard layout, open with ample negative space; a flat bluestone clearing in the center for dancing; clean, uncluttered background.\nLighting: soft early-morning natural daylight, cool white with a faint golden tint, slanting into the courtyard from above at the side; the peach tree casts faint dappled shadows on the bluestone slabs; overall bright and clear, soft light-to-dark transitions, a thin morning-mist feel in the air, petals semi-translucent in the light; sweet, gentle, quiet, healing atmosphere.\nImage 1: ENVIRONMENT only, never identity. Keep architecture, materials, furniture and object layout. Ignore character-sheet backgrounds.\nAll cuts/segments, close-up/reverse included: keep geometry, doors/windows, materials, furniture, prop/light positions; off-screen stays off-screen. Reframe within this set. Scripted prop/light changes persist. Empty sets never remove scripted actors/actions.\n[/HB_SCENE_LOCK_V1]\nReference image mapping (visual only): 1=environment <Picture 1> \u53e4\u98ce\u5ead\u9662.\nEnvironment images define architecture, not an empty cast or a shot to insert. Follow the scripted framing and preserve present actors until a motivated exit/cut; never replace them with the empty reference view.\n\nAudio direction: no speech. No dialogue. No narration. Only scripted ambience and nonverbal reactions; no invented words or subtitles.", "width": 1280, "height": 704, "length": ["7", 1], "ref_image_size": "max", "ref_images.ref_image_0": ["canvas_reference_1", 0]}, "_meta": {"title": "MiniMax H3 \u5168\u80fd\u591a\u53c2\u8003\u56fe\u751f\u89c6\u9891"}}, "6": {"class_type": "PrimitiveFloat", "inputs": {"value": 5.0}, "_meta": {"title": "\u65f6\u957f\uff08\u79d2\uff09"}}, "7": {"class_type": "ComfyMathExpression", "inputs": {"expression": "max(5, round(a * 24)) + (5 - (max(5, round(a * 24)) % 17)) % 17", "values.a": ["6", 0]}, "_meta": {"title": "H3 \u5e27\u6570\u5bf9\u9f50\uff0824fps\uff0c17k+5\uff09"}}, "8": {"class_type": "RandomNoise", "inputs": {"noise_seed": 4652457024961086360}, "_meta": {"title": "\u968f\u673a\u79cd\u5b50"}}, "9": {"class_type": "MiniMaxH3MemoryEfficientSageAttentionPatch", "inputs": {"model": ["1", 0]}, "_meta": {"title": "MiniMax H3 SageAttention \u663e\u5b58\u4f18\u5316"}}, "10": {"class_type": "LoraLoaderModelOnly", "inputs": {"model": ["9", 0], "lora_name": "minimax_h3_turbo_4\u6b65\u52a0\u901f_comfyui.safetensors", "strength_model": 1.0}, "_meta": {"title": "MiniMax H3 4\u6b65\u52a0\u901f LoRA"}}, "11": {"class_type": "BasicScheduler", "inputs": {"model": ["10", 0], "scheduler": "beta", "steps": 8, "denoise": 1.0}, "_meta": {"title": "MiniMax H3 Turbo \u8c03\u5ea6\u5668"}}, "12": {"class_type": "KSamplerSelect", "inputs": {"sampler_name": "euler"}, "_meta": {"title": "MiniMax H3 Turbo \u91c7\u6837\u5668"}}, "13": {"class_type": "BasicGuider", "inputs": {"model": ["10", 0], "conditioning": ["5", 0]}, "_meta": {"title": "MiniMax H3 \u5f15\u5bfc"}}, "14": {"class_type": "SamplerCustomAdvanced", "inputs": {"noise": ["8", 0], "guider": ["13", 0], "sampler": ["12", 0], "sigmas": ["11", 0], "latent_image": ["5", 1]}, "_meta": {"title": "MiniMax H3 Turbo \u9ad8\u7ea7\u91c7\u6837"}}, "15": {"class_type": "VAEDecode", "inputs": {"samples": ["14", 0], "vae": ["3", 0]}, "_meta": {"title": "\u89e3\u7801\u89c6\u9891"}}, "16": {"class_type": "VAEDecodeAudio", "inputs": {"samples": ["14", 0], "vae": ["4", 0]}, "_meta": {"title": "\u89e3\u7801\u539f\u751f\u97f3\u9891"}}, "17": {"class_type": "CreateVideo", "inputs": {"images": ["15", 0], "audio": ["16", 0], "fps": 24.0, "bit_depth": 8}, "_meta": {"title": "\u5408\u6210 H3 \u97f3\u89c6\u9891"}}, "18": {"class_type": "SaveVideo", "inputs": {"video": ["17", 0], "filename_prefix": "video/MiniMax_H3_AllReference_4step_canvas-c8e7e771db0dff7b", "format": "auto", "codec": "auto"}, "_meta": {"title": "\u4fdd\u5b58 MiniMax H3 \u5168\u80fd\u53c2\u8003\u89c6\u9891"}}, "canvas_reference_1": {"class_type": "LoadImage", "inputs": {"image": "canvas-c8e7e771db0dff7b-reference-1.png"}, "_meta": {"title": "Canvas reference 1"}}}, {"client_id": "canvas-c8e7e771db0dff7b", "create_time": 1795085198757}, ["18"]], "outputs": {}, "status": {"status_str": "error", "completed": false, "messages": [["execution_start", {"prompt_id": "a40ef771-5ec1-4b65-816d-41ec318e60aa", "timestamp": 1795085198758}], ["execution_cached", {"nodes": ["1", "2", "3", "4", "6", "7", "9", "10", "11", "12"], "prompt_id": "a40ef771-5ec1-4b65-816d-41ec318e60aa", "timestamp": 1795085198774}], ["execution_error", {"prompt_id": "a40ef771-5ec1-4b65-816d-41ec318e60aa", "node_id": "14", "node_type": "SamplerCustomAdvanced", "executed": ["5", "8", "13", "canvas_reference_1"], "exception_message": "Allocation on device 0 would exceed allowed memory. (out of memory)\nCurrently allocated : 62.03 GiB\nRequested : 992.25 MiB\nDevice limit : 83.05 GiB\nFree (according to CUDA): 4.62 MiB\nPyTorch limit (set by user-supplied memory fraction)\n : 17179869184.00 GiB\nThis error means you ran out of memory on your GPU.\n\nTIPS: If the workflow worked before you might have accidentally set the batch_size to a large number.", "exception_type": "torch.OutOfMemoryError", "traceback": [" File \"/app/execution.py\", line 547, in execute\n output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 352, in get_output_data\n return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/execution.py\", line 326, in _async_map_node_over_list\n await process_inputs(input_dict, i)\n", " File \"/app/execution.py\", line 314, in process_inputs\n result = f(**inputs)\n ^^^^^^^^^^^\n", " File \"/app/comfy_api/internal/__init__.py\", line 149, in wrapped_func\n return method(locked_class, **inputs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_api/latest/_io.py\", line 2053, in EXECUTE_NORMALIZED\n to_return = cls.execute(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy_extras/nodes_custom_sampler.py\", line 1055, in execute\n samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1335, in sample\n output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1262, in outer_sample\n output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1237, in inner_sample\n samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1005, in sample\n samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/utils/_contextlib.py\", line 124, in decorate_context\n return func(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/k_diffusion/sampling.py\", line 205, in sample_euler\n denoised = model(x, sigma_hat * s_in, **extra_args)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 640, in __call__\n out = self.inner_model(x, sigma, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1208, in __call__\n return self.outer_predict_noise(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1215, in outer_predict_noise\n ).execute(x, timestep, model_options, seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 1218, in predict_noise\n return sampling_function(self.inner_model, x, timestep, self.conds.get(\"negative\", None), self.conds.get(\"positive\", None), self.cfg, model_options=model_options, seed=seed)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 620, in sampling_function\n out = calc_cond_batch(model, conds, x, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 211, in calc_cond_batch\n return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 219, in _calc_cond_batch_outer\n return executor.execute(model, conds, x_in, timestep, model_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/samplers.py\", line 335, in _calc_cond_batch\n output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 210, in apply_model\n return comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/model_base.py\", line 254, in _apply_model\n model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 577, in forward\n graph_out = comfy.patcher_extension.WrapperExecutor.new_class_executor(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/patcher_extension.py\", line 113, in execute\n return self.original(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 764, in _forward\n h = block(h, t_emb, mod_segments, rope_freqs, transformer_options=transformer_options)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 293, in forward\n shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = self.adaln_proj(t_emb)\n ^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ldm/minimax/model.py\", line 225, in forward\n x = self.linear(nn.functional.silu(t_emb) if self.apply_silu else t_emb)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1779, in _wrapped_call_impl\n return self._call_impl(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py\", line 1790, in _call_impl\n return forward_call(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 1473, in forward\n output = self.forward_comfy_cast_weights(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 1406, in forward_comfy_cast_weights\n with CastBiasWeightContext(\n ^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 468, in __init__\n self.state = (None, None) if self.slf is None else cast_bias_weight(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 382, in cast_bias_weight\n weight, bias = resolve_cast_module_with_vbar(s, dtype, device, bias_dtype, compute_dtype, want_requant)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 329, in resolve_cast_module_with_vbar\n weight = post_cast(s, \"weight\", weight, dtype, prefetch[\"resident\"], update_weight)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 312, in post_cast\n x = to_dequant(x, dtype if compute_dtype is None else compute_dtype)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/app/comfy/ops.py\", line 306, in to_dequant\n tensor = tensor.dequantize()\n ^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/comfy_kitchen/tensor/base.py\", line 291, in dequantize\n full = self.layout_cls.dequantize(qdata, self._params)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/comfy_kitchen/tensor/int8.py\", line 172, in dequantize\n result = torch.ops.comfy_kitchen.dequantize_int8_convrot_weight_dtype(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/_ops.py\", line 1269, in __call__\n return self._op(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/_library/custom_ops.py\", line 347, in backend_impl\n result = self._backend_fns[device_type](*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/_compile.py\", line 54, in inner\n return disable_fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/_dynamo/eval_frame.py\", line 1263, in _fn\n return fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/torch/_library/custom_ops.py\", line 382, in wrapped_fn\n return fn(*args, **kwargs)\n ^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/comfy_kitchen/backends/eager/quantization.py\", line 1163, in _op_dequantize_int8_convrot_weight_dtype\n return impl(**kwargs)\n ^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/comfy_kitchen/backends/cuda/__init__.py\", line 1699, in dequantize_int8_convrot_weight_dtype\n return _rotate_weight(dequantize_int8_simple(q_2d, scale), h, group_size).to(output_dtype)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n", " File \"/opt/conda/lib/python3.11/site-packages/comfy_kitchen/tensor/int8_utils.py\", line 54, in _rotate_weight\n weight_rotated = torch.matmul(weight_grouped, h_t)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"], "current_inputs": {"noise": ["</app/comfy_extras/nodes_custom_sampler.Noise_RandomNoise object at 0x75a728907a90>"], "guider": ["</app/comfy_extras/nodes_custom_sampler.Guider_Basic object at 0x75a4ccff4990>"], "sampler": ["<comfy.samplers.KSAMPLER object at 0x75a48da3de10>"], "sigmas": ["tensor([1.0000, 0.9950, 0.9825, 0.9607, 0.9234, 0.8553, 0.7207, 0.4249, 0.0000])"], "latent_image": ["{'samples': NestedTensor([tensor([[[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n ...,\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]],\n\n\n [[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n ...,\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]]]), tensor([[[[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n ...,\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]],\n\n [[0., 0., 0., ..., 0., 0., 0.],\n [0., 0., 0., ..., 0., 0., 0.]]]])])}"]}, "current_outputs": ["8", "2", "3", "10", "1", "4", "13", "5", "11", "17", "6", "18", "14", "9", "canvas_reference_1", "12", "16", "7", "15"], "timestamp": 1795085219402}]]}, "meta": {}}} |