Sign In

OpenAI meets Flux [ComfyUI Workflow]

33
260
13
Updated: Apr 25, 2025
toolcomfyuiopenaiupscalingapi
Type
Workflows
Stats
260
0
Reviews
Published
Apr 24, 2025
Base Model
Flux.1 D
Hash
AutoV2
2F86617E7E
Happy Mushroom
denrakeiw
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

OpenAI meets Flux [ComfyUI Workflow]

  1. An initial image is generated using OpenAI GPT Image1 via the ComfyUI API.

  2. The generated image is described/analyzed by a local Ollama LLaVA VLM, which creates detailed prompts specifically for upscaling.

  3. These prompts are encoded with CLIP and T5XXL for optimal conditioning.

  4. The image is then upscaled and enhanced using the Flux Ultimate SD Upscaler and Flux models, with advanced detail restoration and block-busting.

  5. The final high-resolution, high-quality image is saved and can be compared to the original for quality assessment.

This workflow automates prompt creation and upscaling, ensuring the best possible image quality with minimal manual input.