Are you a ComfyUI user looking to take your image generation game to the next level? Check out the latest updates to the ComfyUI IF AI Tools repository (https://github.com/if-ai/ComfyUI-IF_AI_tools)!
Following on my IF_prompt_MKR extension for A1111, Forge and SDnext. I decided to port it to ComfyUI but this time with more tools it will grow into a full fledged AI toolbox
You can use it to generate high quality imades harnessing fast local LLM serving with Ollama or via API from Anthropic and Openai more comming soon
Do you want to reverse engineer a prompt feed an image to the image 2 prompt node and have the VLmodel give you an SD prompt
It works in batches or individual images
One click image generation
Use local LLMs with Ollama or Use Openai API and Anthropic API
ollama can be running on another computer in your network or in the same computer
You can use fast cheap models like haiku to analyze images or reverse engineer a prompt
As an example the images here are automatically enhance from a short prompt:
Ancient Structure, Lone figure🤯
New Features:
1. Anthropic API Support: Harness the power of Anthropic's advanced language models like Claude-3 (Opus, Sonnet, and Haiku) to generate high-quality prompts and image descriptions.😻
2. OpenAI API Integration: Leverage the cutting-edge capabilities of OpenAI's GPT-4 and GPT-3.5 models, including the vision-enabled variants, for enhanced image understanding and prompt generation.
3. Local LLM Utilization: Leverage the versatility of local Large Language Models (LLMs) via Ollama for seamless and efficient prompt generation and image analysis.
4. Vision Model Excellence: Experience superior image-to-prompt capabilities with the inclusion of models like Haiku, designed specifically for visual understanding tasks.
With these exciting updates, the ComfyUI IF AI Tools repository offers a comprehensive suite of tools to streamline your image generation workflow:
- IF Prompt to Prompt Node: Generate compelling and descriptive prompts tailored to your subject matter using local LLMs or advanced APIs like Anthropic and OpenAI.
- IF Image to Prompt Node: Analyze and understand your images, extracting vivid and detailed prompts that encapsulate the visual elements, perfect for creating variations or guiding future generations.👀
- IF Save Text Node: Easily save generated text, prompts, and descriptions in various formats (CSV, TXT, JSON) for future reference or analysis.
- IF Display Text Node: Conveniently display generated text within ComfyUI for quick review and evaluation.
Don't miss out on this powerful toolkit! Head over to the repository (https://github.com/if-ai/ComfyUI-IF_AI_tools), star it, and start exploring the incredible possibilities of combining advanced language and vision models with your ComfyUI workflows.
Let your creativity soar with the ComfyUI IF AI Tools!