Couldn't you also take the CLIP Text Encode node and convert the text into an Input and then just pipe the GripTape output into this? Not that what you're demonstrating is bad or wrong, just another option.
That's a great suggestion - I hadn't thought of it! You absolutely could do this. For those of you who are interested, here are the steps: 1. Right Mouse Button over the CLIP Text Encode (Prompt) node and choose Convert Widget to Input -> Convert Text to Input. Note: Make sure you don't click with the RMB over the textarea - it won't work. you need to click over the node. 2. Connect the output of the agent into the new input node. Simple-as! Great suggestion. :)
Sure, it was this: Output a concise list of requirements for an image generation prompt using the following characteristics. Always include the following: * Subject * Attributes of the -subject(ex: looking to the left, funny expression, detailed interior) * visual characteristics of the image * Camera angle * camera shot type * lighting (accentuate the subject) * art style * color scheme * surrounding environment * foreground elements (blurry leaves, dark shadows) * descriptors of quality * camera used (ex: Nikon d850 film stock, cinestill 800 tungston, polaroid) Prompts should be a comma separated list with the most important elements first. Examples: small puppy, extremely furry, golden hour, low angle, cinematic lighting directed towards the eyes, slight vignette, realistic photograph, foreground leaves, reds and blues, out in nature, best quality girl, looking to the side, jeweled crwon, close-up, cinematic lighting accentuating the girl, fantasy art, digital painting, ethereal glow, vibrant colors, forest, best quality,m masterpiece ((masterpiece)), (top quality), (best quality), girl looking to the side, jeweled crown, close-up, cinematic lighting, vibrant colors, fantasy art, drawn with crayons, ethereal glow, forest Reply
![PASULOL รามเกียรติ์ ตอนที่ 4 นนทกพบรัก [Ramakien Ep.4: Nonthok in love]](http://i.ytimg.com/vi/tMUQgmZzf_A/mqdefault.jpg)
Couldn't you also take the CLIP Text Encode node and convert the text into an Input and then just pipe the GripTape output into this? Not that what you're demonstrating is bad or wrong, just another option.
That's a great suggestion - I hadn't thought of it!
You absolutely could do this. For those of you who are interested, here are the steps:
1. Right Mouse Button over the CLIP Text Encode (Prompt) node and choose Convert Widget to Input -> Convert Text to Input. Note: Make sure you don't click with the RMB over the textarea - it won't work. you need to click over the node.
2. Connect the output of the agent into the new input node.
Simple-as! Great suggestion. :)
thanks for the feedback!
Thank you, thank you, thank you. Why i missed the clip part I havent a clue
You're welcome
could you give me the rules text detail?
Sure, it was this:
Output a concise list of requirements for an image generation prompt using the following characteristics.
Always include the following:
* Subject
* Attributes of the -subject(ex: looking to the left, funny expression, detailed interior)
* visual characteristics of the image
* Camera angle
* camera shot type
* lighting (accentuate the subject)
* art style
* color scheme
* surrounding environment
* foreground elements (blurry leaves, dark shadows)
* descriptors of quality
* camera used (ex: Nikon d850 film stock, cinestill 800 tungston, polaroid)
Prompts should be a comma separated list with the most important elements first.
Examples:
small puppy, extremely furry, golden hour, low angle, cinematic lighting directed towards the eyes, slight vignette, realistic photograph, foreground leaves, reds and blues, out in nature, best quality
girl, looking to the side, jeweled crwon, close-up, cinematic lighting accentuating the girl, fantasy art, digital painting, ethereal glow, vibrant colors, forest, best quality,m masterpiece
((masterpiece)), (top quality), (best quality), girl looking to the side, jeweled crown, close-up, cinematic lighting, vibrant colors, fantasy art, drawn with crayons, ethereal glow, forest
Reply
@@griptape-ai thank you very much