Wan 2.1 Text to Video

# Python 3.10+: pip install comfy-sdk==0.4.0
import os

from comfy_sdk import Comfy

api_key = os.environ.get("COMFY_API_KEY", "").strip()
if not api_key: raise RuntimeError("Set COMFY_API_KEY")
os.environ.setdefault("COMFY_BASE_URL", "https://cloud.comfy.org")
client = Comfy(api_key=api_key)
workflow = client.workflows.from_json({
    "3": {
        "class_type": "KSampler",
        "inputs": {
            "cfg": 6,
            "denoise": 1,
            "latent_image": [
                "40",
                0
            ],
            "model": [
                "48",
                0
            ],
            "negative": [
                "7",
                0
            ],
            "positive": [
                "6",
                0
            ],
            "sampler_name": "uni_pc",
            "scheduler": "simple",
            "seed": 82628696717253,
            "steps": 30
        },
        "_meta": {
            "title": "KSampler"
        }
    },
    "6": {
        "class_type": "CLIPTextEncode",
        "inputs": {
            "clip": [
                "38",
                0
            ],
            "text": "A sweeping cinematic tracking shot glides low across a frost-bound Pleistocene glacial basin during freezing lavender dusk. Two shaggy woolly mammoths charge in full gallop across the vast, wind-scoured permafrost, kicking up dense plumes of crystalline snow with each thundering stride. The camera pans parallel alongside their lumbering sprint, rendering sharp motion blur in the churning icy foreground while maintaining crisp focus on their matted dark brown pelts, curved ivory tusks, and steamy breath billowing into the sub-zero air. Beyond the churning drifts, the palette shifts into deep slate grays, cold cobalt blues, and pale silver reflections across jagged glaciers and looming basalt cliffs stretching endlessly toward the frozen horizon."
        },
        "_meta": {
            "title": "CLIP Text Encode (Positive Prompt)"
        }
    },
    "7": {
        "class_type": "CLIPTextEncode",
        "inputs": {
            "clip": [
                "38",
                0
            ],
            "text": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走"
        },
        "_meta": {
            "title": "CLIP Text Encode (Negative Prompt)"
        }
    },
    "8": {
        "class_type": "VAEDecode",
        "inputs": {
            "samples": [
                "3",
                0
            ],
            "vae": [
                "39",
                0
            ]
        },
        "_meta": {
            "title": "VAE Decode"
        }
    },
    "37": {
        "class_type": "UNETLoader",
        "inputs": {
            "unet_name": "wan2.1_t2v_1.3B_fp16.safetensors",
            "weight_dtype": "default"
        },
        "_meta": {
            "title": "Load Diffusion Model"
        }
    },
    "38": {
        "class_type": "CLIPLoader",
        "inputs": {
            "clip_name": "umt5_xxl_fp8_e4m3fn_scaled.safetensors",
            "device": "default",
            "type": "wan"
        },
        "_meta": {
            "title": "Load CLIP"
        }
    },
    "39": {
        "class_type": "VAELoader",
        "inputs": {
            "vae_name": "wan_2.1_vae.safetensors"
        },
        "_meta": {
            "title": "Load VAE"
        }
    },
    "40": {
        "class_type": "EmptyHunyuanLatentVideo",
        "inputs": {
            "batch_size": 1,
            "height": 480,
            "length": 81,
            "width": 832
        },
        "_meta": {
            "title": "Empty Hunyuan Video 1.0 Latent"
        }
    },
    "48": {
        "class_type": "ModelSamplingSD3",
        "inputs": {
            "model": [
                "37",
                0
            ],
            "shift": 8
        },
        "_meta": {
            "title": "ModelSamplingSD3"
        }
    },
    "49": {
        "class_type": "CreateVideo",
        "inputs": {
            "bit_depth": "auto",
            "codec": "none",
            "color_space": "sRGB",
            "fps": 16,
            "images": [
                "8",
                0
            ]
        },
        "_meta": {
            "title": "Create Video"
        }
    },
    "50": {
        "class_type": "SaveVideo",
        "inputs": {
            "filename_prefix": "video/ComfyUI",
            "format": "auto",
            "format.codec": "auto",
            "video": [
                "49",
                0
            ]
        },
        "_meta": {
            "title": "Save Video"
        }
    }
})

# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)
for output in job.get_outputs("50"):
    output.to_file(output.name)