{ "test_id": "ZRF-I2I-01", "status": "completed", "prompt_id": "ac1d835c-659e-4328-909d-b24ceb15343c", "server": "http://192.168.1.202:8188", "input_image": "refiner_reference.jpg", "input_image_sha256": "f478ab8064bd0642401cbf6bab9aa95380ed4c05d50c6d2959bc9a0eb632354a", "prompt_source": "../I2I/qwen/workflow.json#7", "prompt_used_by_base": true, "prompt_used_by_refiner": true, "base_stage": { "model": "qwen_image_2.1_int8_convrot.safetensors", "text_encoder": "qwen3vl_8b_w4a8.safetensors", "vae": "qwen_image_2.1_vae_bf16.safetensors", "generation_width": 576, "generation_height": 1024, "steps": 8, "cfg": 1.0, "sampler": "euler_ancestral", "scheduler": "beta", "denoise": 1.0 }, "refiner_stage": { "model": "z_image_turbo_int8_convrot.safetensors", "text_encoder": "uncensoredQwen34BLLMForZ_fp8.safetensors", "vae": "ae.safetensors", "internal_width": 1088, "internal_height": 1920, "steps": 5, "cfg": 1.0, "sampler": "dpmpp_2m_sde", "scheduler": "beta", "denoise": 0.33, "sampling_shift": 3.0 }, "generation_time_seconds": 33.571, "outputs": { "base": { "path": "results/ZRF-I2I-01-base.png", "width": 1080, "height": 1920, "sha256": "f6161ea1edb31ed9a85254a2360f61f102248348243e7f958bf17cd13dd18021" }, "refined": { "path": "results/ZRF-I2I-01.png", "width": 1080, "height": 1920, "sha256": "a3751a19d49ced3df4c624bd7dcab430387674c7139286502b44abd2c29a84dc" } }, "review": "Successful ZImage refiner run. The first attempt used TextEncodeZImageOmni with the base image as a second reference and produced a visible grid/ghosting artifact. The final run uses the standard ZImage image-to-image path: the same canonical prompt plus the Qwen base latent, with an internal 1088x1920 aligned canvas and a final 1080x1920 resize. The refined image preserves the subject, glasses, hair, suit, and portrait framing without the artifact." }