Upload ta_workflow (62).json

LLM：
You are a professional meme generation prompt optimizer, specializing in converting any uploaded content into a 3x3 grid of meme expression packs, with strict adherence to the following steps and format:
Step 1: From the uploaded content, extract only the core subject (e.g., animal, character, object, etc.) and its key visual features (such as appearance, color, shape, original basic style if distinguishable, etc.). Do not add guessed or non-existent features; only rely on the actual content of the uploaded material.
Step 2: Rewrite a new image generation prompt that transforms the extracted core subject into a 3x3 grid meme set, following these strict constraints:
The final output must be a 3x3 grid (nine independent images) of meme expression packs, without exception.
Each of the nine images must depict the core subject with distinct, vivid expressions and playful gestures (e.g., smiling, pouting, winking, surprised, angry, cute, etc.), ensuring no repetition of expressions/gestures.
Among the nine images, there must be at least one image in an anime style (the anime style should match the core subject's characteristics to maintain visual coordination), while the remaining eight images can be in a cute, engaging style consistent with the subject's basic attributes.
Keep the core subject's essential visual features (extracted in Step 1) unchanged; do not alter its core shape, main color, or recognizable characteristics.
Do not change the overall tone of the uploaded content (e.g., if the original is warm-toned, the meme set should remain warm-toned; if the original is minimalist, the meme set should stay minimalist) unless the tone conflicts with the "cute, engaging" requirement of meme expression packs.
All nine images should be suitable for expressive online communication (e.g., for social media comments, chat interactions, etc.).
Use this format only:
"[Extracted core subject + its key visual features] transformed into a 3x3 grid meme set. The set includes nine images: each image shows the subject with distinct vivid expressions and playful gestures, with at least one image in anime style; all images retain the subject's essential visual features, maintain the original content's overall tone (if consistent with cute style), are in a cute and engaging style, and are suitable for online communication."

ta_workflow (62).json

@@ -0,0 +1,1325 @@
+{
+  "id": "bcfaae85-1d85-4d94-8a10-22061aa03987",
+  "revision": 0,
+  "last_node_id": 82,
+  "last_link_id": 88,
+  "nodes": [
+    {
+      "id": 42,
+      "type": "ReferenceLatent",
+      "pos": [
+        1340,
+        130
+      ],
+      "size": [
+        211.60000610351562,
+        46
+      ],
+      "flags": {},
+      "order": 18,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "conditioning",
+          "name": "conditioning",
+          "type": "CONDITIONING",
+          "link": 37
+        },
+        {
+          "label": "latent",
+          "name": "latent",
+          "shape": 7,
+          "type": "LATENT",
+          "link": 38
+        }
+      ],
+      "outputs": [
+        {
+          "label": "CONDITIONING",
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            31
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "ReferenceLatent"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 59,
+      "type": "CheckpointLoaderSimple",
+      "pos": [
+        -636,
+        630
+      ],
+      "size": [
+        270,
+        122
+      ],
+      "flags": {},
+      "order": 0,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "MODEL",
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            52
+          ]
+        },
+        {
+          "label": "CLIP",
+          "name": "CLIP",
+          "type": "CLIP",
+          "links": null
+        },
+        {
+          "label": "VAE",
+          "name": "VAE",
+          "type": "VAE",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CheckpointLoaderSimple"
+      },
+      "widgets_values": [
+        "879112449935019302",
+        "FLUX.1 Kontext Dev - fp8"
+      ]
+    },
+    {
+      "id": 39,
+      "type": "VAEEncode",
+      "pos": [
+        1012,
+        304
+      ],
+      "size": [
+        140,
+        46
+      ],
+      "flags": {},
+      "order": 7,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "pixels",
+          "name": "pixels",
+          "type": "IMAGE",
+          "link": 50
+        },
+        {
+          "label": "vae",
+          "name": "vae",
+          "type": "VAE",
+          "link": 35
+        }
+      ],
+      "outputs": [
+        {
+          "label": "LATENT",
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            38
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "VAEEncode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 11,
+      "type": "DualCLIPLoader",
+      "pos": [
+        100,
+        318
+      ],
+      "size": [
+        270,
+        130
+      ],
+      "flags": {},
+      "order": 1,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "CLIP",
+          "name": "CLIP",
+          "type": "CLIP",
+          "slot_index": 0,
+          "links": [
+            54
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "DualCLIPLoader"
+      },
+      "widgets_values": [
+        "t5xxl_fp8_e4m3fn.safetensors",
+        "clip_l.safetensors",
+        "flux",
+        "default"
+      ]
+    },
+    {
+      "id": 10,
+      "type": "VAELoader",
+      "pos": [
+        1652,
+        -178
+      ],
+      "size": [
+        270,
+        58
+      ],
+      "flags": {},
+      "order": 2,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "VAE",
+          "name": "VAE",
+          "type": "VAE",
+          "slot_index": 0,
+          "links": [
+            22,
+            35
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "VAELoader"
+      },
+      "widgets_values": [
+        "ae.safetensors"
+      ]
+    },
+    {
+      "id": 22,
+      "type": "BasicGuider",
+      "pos": [
+        1583.4166259765625,
+        130
+      ],
+      "size": [
+        161.1999969482422,
+        46
+      ],
+      "flags": {},
+      "order": 19,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "model",
+          "name": "model",
+          "type": "MODEL",
+          "link": 30
+        },
+        {
+          "label": "conditioning",
+          "name": "conditioning",
+          "type": "CONDITIONING",
+          "link": 31
+        }
+      ],
+      "outputs": [
+        {
+          "label": "GUIDER",
+          "name": "GUIDER",
+          "type": "GUIDER",
+          "slot_index": 0,
+          "links": [
+            25
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "BasicGuider"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 16,
+      "type": "KSamplerSelect",
+      "pos": [
+        100,
+        578
+      ],
+      "size": [
+        270,
+        58
+      ],
+      "flags": {},
+      "order": 3,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "SAMPLER",
+          "name": "SAMPLER",
+          "type": "SAMPLER",
+          "slot_index": 0,
+          "links": [
+            26
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "KSamplerSelect"
+      },
+      "widgets_values": [
+        "euler_ancestral"
+      ]
+    },
+    {
+      "id": 13,
+      "type": "SamplerCustomAdvanced",
+      "pos": [
+        1762,
+        101
+      ],
+      "size": [
+        236.8000030517578,
+        106
+      ],
+      "flags": {},
+      "order": 20,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "noise",
+          "name": "noise",
+          "type": "NOISE",
+          "link": 24
+        },
+        {
+          "label": "guider",
+          "name": "guider",
+          "type": "GUIDER",
+          "link": 25
+        },
+        {
+          "label": "sampler",
+          "name": "sampler",
+          "type": "SAMPLER",
+          "link": 26
+        },
+        {
+          "label": "sigmas",
+          "name": "sigmas",
+          "type": "SIGMAS",
+          "link": 27
+        },
+        {
+          "label": "latent_image",
+          "name": "latent_image",
+          "type": "LATENT",
+          "link": 28
+        }
+      ],
+      "outputs": [
+        {
+          "label": "output",
+          "name": "output",
+          "type": "LATENT",
+          "links": [
+            21
+          ]
+        },
+        {
+          "label": "denoised_output",
+          "name": "denoised_output",
+          "type": "LATENT",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "SamplerCustomAdvanced"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 8,
+      "type": "VAEDecode",
+      "pos": [
+        2103.316650390625,
+        130
+      ],
+      "size": [
+        140,
+        46
+      ],
+      "flags": {},
+      "order": 21,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "samples",
+          "name": "samples",
+          "type": "LATENT",
+          "link": 21
+        },
+        {
+          "label": "vae",
+          "name": "vae",
+          "type": "VAE",
+          "link": 22
+        }
+      ],
+      "outputs": [
+        {
+          "label": "IMAGE",
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "slot_index": 0,
+          "links": [
+            62
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "VAEDecode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 30,
+      "type": "ModelSamplingFlux",
+      "pos": [
+        616,
+        413
+      ],
+      "size": [
+        270,
+        130
+      ],
+      "flags": {},
+      "order": 12,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "model",
+          "name": "model",
+          "type": "MODEL",
+          "link": 53
+        },
+        {
+          "label": "width",
+          "name": "width",
+          "type": "INT",
+          "widget": {
+            "name": "width"
+          },
+          "link": 81
+        },
+        {
+          "label": "height",
+          "name": "height",
+          "type": "INT",
+          "widget": {
+            "name": "height"
+          },
+          "link": 82
+        }
+      ],
+      "outputs": [
+        {
+          "label": "MODEL",
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            30,
+            51
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "ModelSamplingFlux"
+      },
+      "widgets_values": [
+        1.15,
+        0.5,
+        1024,
+        1024
+      ]
+    },
+    {
+      "id": 6,
+      "type": "CLIPTextEncode",
+      "pos": [
+        470,
+        127
+      ],
+      "size": [
+        400,
+        200
+      ],
+      "flags": {},
+      "order": 16,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "clip",
+          "name": "clip",
+          "type": "CLIP",
+          "link": 54
+        },
+        {
+          "label": "text",
+          "name": "text",
+          "type": "STRING",
+          "widget": {
+            "name": "text"
+          },
+          "link": 87
+        }
+      ],
+      "outputs": [
+        {
+          "label": "CONDITIONING",
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            32
+          ]
+        }
+      ],
+      "title": "CLIP Text Encode (Positive Prompt)",
+      "properties": {
+        "Node name for S&R": "CLIPTextEncode"
+      },
+      "widgets_values": [
+        ""
+      ]
+    },
+    {
+      "id": 76,
+      "type": "GenerateCaptionAdvanceNode",
+      "pos": [
+        -72,
+        -656
+      ],
+      "size": [
+        400,
+        200
+      ],
+      "flags": {},
+      "order": 8,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "images",
+          "name": "images",
+          "type": "IMAGE",
+          "link": 74
+        }
+      ],
+      "outputs": [
+        {
+          "label": "STRING",
+          "name": "STRING",
+          "type": "STRING",
+          "slot_index": 0,
+          "links": [
+            76
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "GenerateCaptionAdvanceNode"
+      },
+      "widgets_values": [
+        "joycaption2",
+        "Describe the image in as much detail as possible using an image generation prompt style.\n"
+      ]
+    },
+    {
+      "id": 27,
+      "type": "EmptySD3LatentImage",
+      "pos": [
+        526,
+        854
+      ],
+      "size": [
+        270,
+        106
+      ],
+      "flags": {},
+      "order": 11,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "width",
+          "name": "width",
+          "type": "INT",
+          "widget": {
+            "name": "width"
+          },
+          "link": 79
+        },
+        {
+          "label": "height",
+          "name": "height",
+          "type": "INT",
+          "widget": {
+            "name": "height"
+          },
+          "link": 80
+        }
+      ],
+      "outputs": [
+        {
+          "label": "LATENT",
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            28
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "EmptySD3LatentImage"
+      },
+      "widgets_values": [
+        1024,
+        1360,
+        1
+      ]
+    },
+    {
+      "id": 81,
+      "type": "GetImageSize+",
+      "pos": [
+        44.849735260009766,
+        968.144287109375
+      ],
+      "size": [
+        158.27890014648438,
+        66
+      ],
+      "flags": {},
+      "order": 9,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "image",
+          "name": "image",
+          "type": "IMAGE",
+          "link": 83
+        }
+      ],
+      "outputs": [
+        {
+          "label": "width",
+          "name": "width",
+          "type": "INT",
+          "links": [
+            79,
+            81
+          ]
+        },
+        {
+          "label": "height",
+          "name": "height",
+          "type": "INT",
+          "links": [
+            80,
+            82
+          ]
+        },
+        {
+          "label": "count",
+          "name": "count",
+          "type": "INT",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "GetImageSize+"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 75,
+      "type": "SaveImage",
+      "pos": [
+        2627.711181640625,
+        -344.45477294921875
+      ],
+      "size": [
+        315,
+        270
+      ],
+      "flags": {},
+      "order": 23,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "images",
+          "name": "images",
+          "type": "IMAGE",
+          "link": 84
+        }
+      ],
+      "outputs": [],
+      "properties": {},
+      "widgets_values": [
+        "TensorArt"
+      ]
+    },
+    {
+      "id": 79,
+      "type": "ShowText|pysssss",
+      "pos": [
+        423.9599609375,
+        -615.07763671875
+      ],
+      "size": [
+        1008.7449340820312,
+        221.1501922607422
+      ],
+      "flags": {},
+      "order": 13,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "text",
+          "name": "text",
+          "type": "STRING",
+          "link": 78
+        }
+      ],
+      "outputs": [
+        {
+          "label": "STRING",
+          "name": "STRING",
+          "shape": 6,
+          "type": "STRING",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "ShowText|pysssss"
+      },
+      "widgets_values": [
+        "<think>\nOkay, let's tackle this query. The user wants me to act as a meme generation prompt optimizer. They provided a specific photo of two cats and need a 3x3 grid of meme expressions based on that. Let me break down the steps.\n\nFirst, Step 1 is extracting the core subject and key visual features. The main subject here is two cats. The details given are: black and white cat on the left, orange tabby on the right. Both sitting on a pink textured surface, white background, electric outlet behind them. They have cute, curious expressions. Specific features include one with a pink nose, green eyes, striped fur, and solid black/white fur. Both looking at the camera, indoor, natural lighting, close-up, no humans, watermark in the corner, jpeg artifacts.\n\nI need to make sure I don't add anything not mentioned. So the core subjects are the two cats with those specific features. Now, moving to Step 2: creating the prompt for a 3x3 grid. Each image must have distinct expressions and gestures, no repeats. At least one anime style, others cute/engaging. Keep original visual features and tone. The original is cute and curious, so the meme set should stay that way unless conflicting. Also, suitable for online communication.\n\nI need to list nine different expressions. Let me brainstorm: smiling, pouting, winking, surprised, angry, playful, sleepy, curious, and maybe one with a gesture like waving. Wait, the user mentioned gestures too, so maybe some have actions. But need to ensure no repetition. Also, one in anime style. The anime style should match the cats' characteristics. The original has a warm tone (pink surface, natural lighting), so keep that.\n\nNow, structure the prompt as specified. Start with the core subject and features, then the set requirements. Make sure to mention the anime style, the expressions, retain features, tone, style, and purpose. Check for any constraints: no altering core shape, color, or recognizable features. Also, the pink textured surface and electric outlet are part of the original, but the prompt focuses on the cats. Wait, the original has them sitting on a pink surface with an outlet behind. But the meme images should focus on the cats. Should those elements be included? The user's instruction says to extract core subject (cats) and their key features. The background elements like pink surface and outlet are part of the original photo but maybe not part of the core subject. The instruction says to extract the core subject (cats) and their key visual features (colors, fur patterns, etc.). So the background elements might not be necessary for the meme prompts. The cats are the main focus.\n\nSo the prompt should focus on the cats' features: two cats, black and white (solid) and orange tabby (striped), pink nose, green eyes, sitting posture. The background can be omitted unless necessary. The original has a white background, but maybe the meme images can have similar settings but focused on expressions.\n\nNow, ensuring that all nine images retain these features. Each image shows the two cats with different expressions. Wait, does the 3x3 grid have both cats in each image, or is it each cat as separate? The original photo has two cats. The core subject is \"two cats\" with specific features. So each meme image should include both cats, maintaining their features. But the expressions and gestures would be per cat? Or each image has both cats with different expressions. For example, one image has both cats smiling, another has one pouting and the other surprised? But the user says each image must have distinct expressions. Wait, the instruction says \"each image shows the subject with distinct vivid expressions and playful gestures\". The subject is two cats. So each image has both cats, each with their own expressions? Or maybe each image focuses on one cat? The original is two cats, so probably each image includes both, but with different expressions and gestures between them. But ensuring that across the nine images, no repetition of expressions. But since there are two cats, maybe each image has two different expressions, but the nine images collectively don't repeat any expression. However, the user says \"each of the nine images must depict the core subject with distinct...\". If the core subject is two cats, then each image has both, but their expressions are unique per image. So for example, Image 1: Cat A smiling, Cat B curious; Image 2: Cat A pouting, Cat B winking; etc. But need to make sure that across all images, each expression is unique. But with two cats per image, it's possible to have two expressions per image. However, the user's instruction says \"no repetition of expressions/gestures\" among the nine images. So each image has unique expressions, but since each image has two cats, maybe each image has"
+      ]
+    },
+    {
+      "id": 17,
+      "type": "BasicScheduler",
+      "pos": [
+        1205,
+        734
+      ],
+      "size": [
+        270,
+        106
+      ],
+      "flags": {},
+      "order": 15,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "model",
+          "name": "model",
+          "type": "MODEL",
+          "link": 51
+        }
+      ],
+      "outputs": [
+        {
+          "label": "SIGMAS",
+          "name": "SIGMAS",
+          "type": "SIGMAS",
+          "slot_index": 0,
+          "links": [
+            27
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "BasicScheduler"
+      },
+      "widgets_values": [
+        "simple",
+        20,
+        1
+      ]
+    },
+    {
+      "id": 71,
+      "type": "CR Upscale Image",
+      "pos": [
+        2316.587158203125,
+        275.6064453125
+      ],
+      "size": [
+        255.25697326660156,
+        222
+      ],
+      "flags": {},
+      "order": 22,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "image",
+          "name": "image",
+          "type": "IMAGE",
+          "link": 62
+        }
+      ],
+      "outputs": [
+        {
+          "label": "IMAGE",
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "slot_index": 0,
+          "links": [
+            84
+          ]
+        },
+        {
+          "label": "show_help",
+          "name": "show_help",
+          "type": "STRING",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CR Upscale Image"
+      },
+      "widgets_values": [
+        "2x-ESRGAN.pth",
+        "rescale",
+        1.5,
+        1024,
+        "lanczos",
+        "true",
+        8
+      ]
+    },
+    {
+      "id": 68,
+      "type": "LoraLoaderModelOnly",
+      "pos": [
+        -241,
+        123
+      ],
+      "size": [
+        315,
+        106
+      ],
+      "flags": {},
+      "order": 6,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "model",
+          "name": "model",
+          "type": "MODEL",
+          "link": 52
+        }
+      ],
+      "outputs": [
+        {
+          "label": "MODEL",
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            53
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "LoraLoaderModelOnly"
+      },
+      "widgets_values": [
+        "908648351619092394",
+        "宠物九宫格 - 1.0",
+        0.9
+      ]
+    },
+    {
+      "id": 25,
+      "type": "RandomNoise",
+      "pos": [
+        100,
+        766
+      ],
+      "size": [
+        270,
+        82
+      ],
+      "flags": {},
+      "order": 4,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "NOISE",
+          "name": "NOISE",
+          "type": "NOISE",
+          "slot_index": 0,
+          "links": [
+            24
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "RandomNoise"
+      },
+      "widgets_values": [
+        799852026427213,
+        "fixed"
+      ]
+    },
+    {
+      "id": 26,
+      "type": "FluxGuidance",
+      "pos": [
+        970,
+        130
+      ],
+      "size": [
+        270,
+        58
+      ],
+      "flags": {},
+      "order": 17,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "conditioning",
+          "name": "conditioning",
+          "type": "CONDITIONING",
+          "link": 32
+        }
+      ],
+      "outputs": [
+        {
+          "label": "CONDITIONING",
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            37
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "FluxGuidance"
+      },
+      "widgets_values": [
+        30
+      ]
+    },
+    {
+      "id": 61,
+      "type": "TensorArt_LoadImage",
+      "pos": [
+        -434.8222961425781,
+        798.8560791015625
+      ],
+      "size": [
+        315,
+        314
+      ],
+      "flags": {},
+      "order": 5,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "label": "IMAGE",
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "slot_index": 0,
+          "links": [
+            50,
+            74,
+            83
+          ]
+        },
+        {
+          "label": "MASK",
+          "name": "MASK",
+          "type": "MASK",
+          "links": null
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "TensorArt_LoadImage"
+      },
+      "widgets_values": [
+        "https://image.tensorartassets.com/workflow_media_daily/user_upload/639195226744978302/40e78164-0ce9-414c-8ea6-8152fd99898c.png",
+        "image"
+      ]
+    },
+    {
+      "id": 80,
+      "type": "TALLMNode",
+      "pos": [
+        424.17559814453125,
+        -1099.3084716796875
+      ],
+      "size": [
+        637.8826293945312,
+        395.802734375
+      ],
+      "flags": {},
+      "order": 10,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "prompt",
+          "name": "prompt",
+          "type": "STRING",
+          "widget": {
+            "name": "prompt"
+          },
+          "link": 76
+        }
+      ],
+      "outputs": [
+        {
+          "label": "STRING",
+          "name": "STRING",
+          "type": "STRING",
+          "slot_index": 0,
+          "links": [
+            78,
+            88
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "TALLMNode"
+      },
+      "widgets_values": [
+        "qwen-3-32b-awq",
+        "You are a professional meme generation prompt optimizer, specializing in converting any uploaded content into a 3x3 grid of meme expression packs, with strict adherence to the following steps and format:\nStep 1: From the uploaded content, extract only the core subject (e.g., animal, character, object, etc.) and its key visual features (such as appearance, color, shape, original basic style if distinguishable, etc.). Do not add guessed or non-existent features; only rely on the actual content of the uploaded material.\nStep 2: Rewrite a new image generation prompt that transforms the extracted core subject into a 3x3 grid meme set, following these strict constraints:\nThe final output must be a 3x3 grid (nine independent images) of meme expression packs, without exception.\nEach of the nine images must depict the core subject with distinct, vivid expressions and playful gestures (e.g., smiling, pouting, winking, surprised, angry, cute, etc.), ensuring no repetition of expressions/gestures.\nAmong the nine images, there must be at least one image in an anime style (the anime style should match the core subject's characteristics to maintain visual coordination), while the remaining eight images can be in a cute, engaging style consistent with the subject's basic attributes.\nKeep the core subject's essential visual features (extracted in Step 1) unchanged; do not alter its core shape, main color, or recognizable characteristics.\nDo not change the overall tone of the uploaded content (e.g., if the original is warm-toned, the meme set should remain warm-toned; if the original is minimalist, the meme set should stay minimalist) unless the tone conflicts with the \"cute, engaging\" requirement of meme expression packs.\nAll nine images should be suitable for expressive online communication (e.g., for social media comments, chat interactions, etc.).\nUse this format only:\n\"[Extracted core subject + its key visual features] transformed into a 3x3 grid meme set. The set includes nine images: each image shows the subject with distinct vivid expressions and playful gestures, with at least one image in anime style; all images retain the subject's essential visual features, maintain the original content's overall tone (if consistent with cute style), are in a cute and engaging style, and are suitable for online communication.\"",
+        "a flower in a vase",
+        0.8,
+        1006
+      ]
+    },
+    {
+      "id": 82,
+      "type": "TextCombinerTwo",
+      "pos": [
+        534.6334228515625,
+        -152.57928466796875
+      ],
+      "size": [
+        270,
+        82
+      ],
+      "flags": {},
+      "order": 14,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "text2",
+          "name": "text2",
+          "type": "STRING",
+          "widget": {
+            "name": "text2"
+          },
+          "link": 88
+        }
+      ],
+      "outputs": [
+        {
+          "label": "text",
+          "name": "text",
+          "type": "STRING",
+          "links": [
+            87
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "TextCombinerTwo"
+      },
+      "widgets_values": [
+        "into a 3x3 grid of meme expression packs",
+        ""
+      ]
+    }
+  ],
+  "links": [
+    [
+      21,
+      13,
+      0,
+      8,
+      0,
+      "LATENT"
+    ],
+    [
+      22,
+      10,
+      0,
+      8,
+      1,
+      "VAE"
+    ],
+    [
+      24,
+      25,
+      0,
+      13,
+      0,
+      "NOISE"
+    ],
+    [
+      25,
+      22,
+      0,
+      13,
+      1,
+      "GUIDER"
+    ],
+    [
+      26,
+      16,
+      0,
+      13,
+      2,
+      "SAMPLER"
+    ],
+    [
+      27,
+      17,
+      0,
+      13,
+      3,
+      "SIGMAS"
+    ],
+    [
+      28,
+      27,
+      0,
+      13,
+      4,
+      "LATENT"
+    ],
+    [
+      30,
+      30,
+      0,
+      22,
+      0,
+      "MODEL"
+    ],
+    [
+      31,
+      42,
+      0,
+      22,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      32,
+      6,
+      0,
+      26,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      35,
+      10,
+      0,
+      39,
+      1,
+      "VAE"
+    ],
+    [
+      37,
+      26,
+      0,
+      42,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      38,
+      39,
+      0,
+      42,
+      1,
+      "LATENT"
+    ],
+    [
+      50,
+      61,
+      0,
+      39,
+      0,
+      "IMAGE"
+    ],
+    [
+      51,
+      30,
+      0,
+      17,
+      0,
+      "MODEL"
+    ],
+    [
+      52,
+      59,
+      0,
+      68,
+      0,
+      "MODEL"
+    ],
+    [
+      53,
+      68,
+      0,
+      30,
+      0,
+      "MODEL"
+    ],
+    [
+      54,
+      11,
+      0,
+      6,
+      0,
+      "CLIP"
+    ],
+    [
+      62,
+      8,
+      0,
+      71,
+      0,
+      "IMAGE"
+    ],
+    [
+      74,
+      61,
+      0,
+      76,
+      0,
+      "IMAGE"
+    ],
+    [
+      76,
+      76,
+      0,
+      80,
+      0,
+      "STRING"
+    ],
+    [
+      78,
+      80,
+      0,
+      79,
+      0,
+      "STRING"
+    ],
+    [
+      79,
+      81,
+      0,
+      27,
+      0,
+      "INT"
+    ],
+    [
+      80,
+      81,
+      1,
+      27,
+      1,
+      "INT"
+    ],
+    [
+      81,
+      81,
+      0,
+      30,
+      1,
+      "INT"
+    ],
+    [
+      82,
+      81,
+      1,
+      30,
+      2,
+      "INT"
+    ],
+    [
+      83,
+      61,
+      0,
+      81,
+      0,
+      "IMAGE"
+    ],
+    [
+      84,
+      71,
+      0,
+      75,
+      0,
+      "IMAGE"
+    ],
+    [
+      87,
+      82,
+      0,
+      6,
+      1,
+      "STRING"
+    ],
+    [
+      88,
+      80,
+      0,
+      82,
+      0,
+      "STRING"
+    ]
+  ],
+  "groups": [],
+  "config": {},
+  "extra": {
+    "ds": {
+      "scale": 0.7247295000000009,
+      "offset": [
+        717.8835844388052,
+        612.1090759936565
+      ]
+    },
+    "frontendVersion": "1.24.1",
+    "groupNodes": {},
+    "VHS_MetadataImage": true,
+    "VHS_latentpreview": false,
+    "VHS_KeepIntermediate": true,
+    "VHS_latentpreviewrate": 0
+  },
+  "version": 0.4
+}