nodes/NeuralSamurAI-ComfyUI-PromptJSON
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update readme.md

Browse Source
This commit is contained in:
NeuralSamurAI
2024-08-07 15:40:39 -04:00
committed by GitHub
parent fc08051f24
commit cd25541124
1 changed files with 8 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
readme.md
+8 -4
View File
@@ -190,9 +190,9 @@ Generate a detailed image description based on the following prompt, adhering to
} }
``` ```
2. Structured Prompt: (The JSON representation of the input prompt, as shown in the LLM Prompt) 2. Structured Prompt: (The JSON representation of the input prompt, as shown in the 2nd half of the LLM Prompt)
3. Custom Schema: (The schema used, as shown at the beginning of the LLM Prompt) 3. Custom / Default Schema: (The schema used, as shown at the beginning of the LLM Prompt)
## Notes ## Notes
- The `structured_prompt` output is an intermediate representation, not the final detailed JSON. - The `structured_prompt` output is an intermediate representation, not the final detailed JSON.
@@ -201,7 +201,7 @@ Generate a detailed image description based on the following prompt, adhering to
- The custom schema defines the structure for both the intermediate prompt and the final LLM output. - The custom schema defines the structure for both the intermediate prompt and the final LLM output.
- Adjust the complexity parameter to control the level of detail in the generated prompts. - Adjust the complexity parameter to control the level of detail in the generated prompts.
## Token Counter Usage ## Token Counter Node Usage
The Token Counter node is available in the "utils" category: The Token Counter node is available in the "utils" category:
1. Connect the following inputs: 1. Connect the following inputs:
@@ -214,4 +214,8 @@ The Token Counter node is available in the "utils" category:
- `clip_token_count`: Number of tokens using CLIP tokenizer (if enabled) - `clip_token_count`: Number of tokens using CLIP tokenizer (if enabled)
## Support ## Support
For issues, feature requests, or contributions, please open an issue or pull request in the GitHub repository. For issues, feature requests, or contributions, please open an issue or pull request in the GitHub repository.
## Acknowledgements
This node builds upon the concept of controllable image generation via JSON prompting, specifically in the Flux1 family of models.
Thanks to the Image Diffusion, AI Twitter, ComfyUI & Banodoco communities for their support and inspiration.