diff --git a/optimized/I2I/qwen-noctq-flux3klein-refiner/README.md b/optimized/I2I/qwen-noctq-flux3klein-refiner/README.md index 6379ebe..eb9f120 100644 --- a/optimized/I2I/qwen-noctq-flux3klein-refiner/README.md +++ b/optimized/I2I/qwen-noctq-flux3klein-refiner/README.md @@ -2,7 +2,9 @@ Two-stage local image-to-image workflow. Qwen NoctQ v4 generates the first image, then the decoded pixels are passed to FLUX.3 Klein as both the image -reference and the VAE-encoded latent for a low-denoise refinement pass. +reference and the VAE-encoded latent for a low-denoise refinement pass. The +FLUX stage has **no prompt text**; it refines only from the supplied image +conditioning and latent. ## Fixed starter settings @@ -18,8 +20,8 @@ is used. The input is `../character_reference.png` at 1080×1920 with seed `310101`. The Qwen output is decoded with the Qwen VAE, then connected to the FLUX -reference conditioner and re-encoded with `flux2-vae.safetensors` before the -0.11-denoise refiner pass. +reference conditioner with an empty prompt and re-encoded with +`flux2-vae.safetensors` before the 0.11-denoise refiner pass. ## Test matrix diff --git a/optimized/I2I/qwen-noctq-flux3klein-refiner/workflow.json b/optimized/I2I/qwen-noctq-flux3klein-refiner/workflow.json index 82f0683..ae97bfc 100644 --- a/optimized/I2I/qwen-noctq-flux3klein-refiner/workflow.json +++ b/optimized/I2I/qwen-noctq-flux3klein-refiner/workflow.json @@ -15,7 +15,7 @@ "14": {"class_type":"UNETLoader","inputs":{"unet_name":"flux3-image-klein-int8-convrot.safetensors","weight_dtype":"default"}}, "15": {"class_type":"CLIPLoader","inputs":{"clip_name":"qwen3-vl-4b-heretic_int8.safetensors","type":"flux3","device":"default"}}, "16": {"class_type":"VAELoader","inputs":{"vae_name":"flux2-vae.safetensors"}}, - "17": {"class_type":"StarFlux2Conditioner","inputs":{"clip":["15",0],"vae":["16",0],"text":"Refine the supplied image while preserving the subject identity, composition, pose, clothing, colors, and overall visual content. Improve realism, detail, anatomy, materials, and lighting without redesigning or adding objects.","join_references":true,"image_1":["13",0]}}, + "17": {"class_type":"StarFlux2Conditioner","inputs":{"clip":["15",0],"vae":["16",0],"text":"","join_references":true,"image_1":["13",0]}}, "18": {"class_type":"FluxGuidance","inputs":{"conditioning":["17",0],"guidance":1.0}}, "19": {"class_type":"VAEEncode","inputs":{"pixels":["13",0],"vae":["16",0]}}, "20": {"class_type":"FluxSampler","inputs":{"model":["14",0],"conditioning":["18",0],"latent_image":["19",0],"sampler_name":"euler_ancestral","scheduler":"simple","steps":8,"denoise":0.11,"noise_seed":310101}},