Has someone setup the sd-server for qwen image edit?

[djismgaming]

Looking to add it under open-webui as I did for the image generation, but this is not working. Like the request does not arrive to the server

C:\Users\Ismael\apps\stablediffusioncpp-rocm\sd-server.exe `
--diffusion-model F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf `
--vae F:\ai\models-image\qwen_image_vae.safetensors `
--llm F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf `
--llm_vision F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf `
--cfg-scale 2.5 --sampling-method euler --diffusion-fa --flow-shift 3 `
--qwen-image-zero-cond-t

I have it behind llama-swap (same as I do successfully for image generation via open-webui) and this is the output of the server. After sending the request via open-webui to "remove the text" and the model changing to ready in llama-swap:

main.cpp:298  - version: stable-diffusion.cpp version unknown, commit d6dd6d7
[DEBUG] main.cpp:299  - System Info: 
    SSE3 = 1 |     AVX = 1 |     AVX2 = 1 |     AVX512 = 0 |     AVX512_VBMI = 0 |     AVX512_VNNI = 0 |     FMA = 1 |     NEON = 0 |     ARM_FMA = 0 |     F16C = 1 |     FP16_VA = 0 |     WASM_SIMD = 0 |     VSX = 0 | 
[DEBUG] main.cpp:300  - SDSvrParams {
  listen_ip: 127.0.0.1,
  listen_port: "5803",
  serve_html_path: "",
}
[DEBUG] main.cpp:301  - SDContextParams {
  n_threads: 6,
  model_path: "",
  clip_l_path: "",
  clip_g_path: "",
  clip_vision_path: "",
  t5xxl_path: "",
  llm_path: "F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf",
  llm_vision_path: "F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf",
  diffusion_model_path: "F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf",
  high_noise_diffusion_model_path: "",
  vae_path: "F:\ai\models-image\qwen_image_vae.safetensors",
  taesd_path: "",
  esrgan_path: "",
  control_net_path: "",
  embedding_dir: "",
  embeddings: {
  }
  wtype: NONE,
  tensor_type_rules: "",
  lora_model_dir: ".",
  photo_maker_path: "",
  rng_type: cuda,
  sampler_rng_type: NONE,
  offload_params_to_cpu: false,
  enable_mmap: false,
  control_net_cpu: false,
  clip_on_cpu: false,
  vae_on_cpu: false,
  flash_attn: false,
  diffusion_flash_attn: true,
  diffusion_conv_direct: false,
  vae_conv_direct: false,
  circular: false,
  circular_x: false,
  circular_y: false,
  chroma_use_dit_mask: true,
  qwen_image_zero_cond_t: true,
  chroma_use_t5_mask: false,
  chroma_t5_mask_pad: 1,
  prediction: NONE,
  lora_apply_mode: auto,
  vae_tiling_params: { 0, 0, 0, 0.5, 0, 0 },
  force_sdxl_vae_conv_scale: false
}
[DEBUG] main.cpp:302  - SDGenerationParams {
  loras: "{
  }",
  high_noise_loras: "{
  }",
  prompt: "",
  negative_prompt: "",
  clip_skip: -1,
  width: -1,
  height: -1,
  batch_count: 1,
  init_image_path: "",
  end_image_path: "",
  mask_image_path: "",
  control_image_path: "",
  ref_image_paths: [],
  control_video_path: "",
  auto_resize_ref_image: true,
  increase_ref_index: false,
  pm_id_images_dir: "",
  pm_id_embed_path: "",
  pm_style_strength: 20,
  skip_layers: [7, 8, 9],
  sample_params: (txt_cfg: 2.50, img_cfg: 2.50, distilled_guidance: 3.50, slg.layer_count: 3, slg.layer_start: 0.01, slg.layer_end: 0.20, slg.scale: 0.00, scheduler: NONE, sample_method: euler, sample_steps: 20, eta: 0.00, shifted_timestep: 0, flow_shift: 3.00),
  high_noise_skip_layers: [7, 8, 9],
  high_noise_sample_params: (txt_cfg: 7.00, img_cfg: 7.00, distilled_guidance: 3.50, slg.layer_count: 3, slg.layer_start: 0.01, slg.layer_end: 0.20, slg.scale: 0.00, scheduler: NONE, sample_method: NONE, sample_steps: 20, eta: 0.00, shifted_timestep: 0, flow_shift: inf),
  custom_sigmas: [],
  cache_mode: "",
  cache_option: "",
  cache: disabled (threshold=1, start=0.15, end=0.95),
  moe_boundary: 0.875,
  video_frames: 1,
  fps: 16,
  vace_strength: 1,
  strength: 0.75,
  control_strength: 0.9,
  seed: 42,
  upscale_repeats: 1,
  upscale_tile_size: 128,
}
[DEBUG] stable-diffusion.cpp:173  - Using CUDA backend
[DEBUG] src\ggml_extend.hpp:78   - ggml_cuda_init: found 1 ROCm devices:
[INFO ] src\ggml_extend.hpp:78   -   Device 0: AMD Radeon RX 6800, gfx1030 (0x1030), VMM: no, Wave Size: 32
[INFO ] stable-diffusion.cpp:267  - loading diffusion model from 'F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf'
[INFO ] model.cpp:366  - load F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf using gguf format
[INFO ] model.cpp:412  - init from 'F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf'
[DEBUG] stable-diffusion.cpp:314  - loading llm from 'F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf'
[INFO ] model.cpp:366  - load F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf using gguf format
[INFO ] model.cpp:412  - init from 'F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf'
[DEBUG] stable-diffusion.cpp:321  - loading llm vision from 'F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf'
[INFO ] model.cpp:366  - load F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf using gguf format
[INFO ] model.cpp:412  - init from 'F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf'
[DEBUG] stable-diffusion.cpp:328  - loading vae from 'F:\ai\models-image\qwen_image_vae.safetensors'
[INFO ] model.cpp:369  - load F:\ai\models-image\qwen_image_vae.safetensors using safetensors format
[INFO ] model.cpp:503  - init from 'F:\ai\models-image\qwen_image_vae.safetensors', prefix = 'vae.'
[DEBUG] stable-diffusion.cpp:345  - Version: Qwen Image 
[INFO ] stable-diffusion.cpp:373  - Weight type stat:                      f32: 1520 |     f16: 34   |    q8_0: 196  |    q4_K: 729  |    q5_K: 20   |    q6_K: 286  |    bf16: 200  
[INFO ] stable-diffusion.cpp:374  - Conditioner weight type stat:          f32: 432  |     f16: 34   |    q8_0: 194  |    q4_K: 169  |    q6_K: 28   
[INFO ] stable-diffusion.cpp:375  - Diffusion model weight type stat:      f32: 1088 |    q8_0: 2    |    q4_K: 560  |    q5_K: 20   |    q6_K: 258  |    bf16: 6    
[INFO ] stable-diffusion.cpp:376  - VAE weight type stat:                 bf16: 194  
[INFO ] stable-diffusion.cpp:378  - ggml tensor size = 400 bytes
[DEBUG] src\llm.hpp:286  - merges size 151387
[DEBUG] src\llm.hpp:318  - vocab size: 151669
[DEBUG] src\llm.hpp:1140 - llm: num_layers = 28, vocab_size = 152064, hidden_size = 3584, intermediate_size = 18944
[DEBUG] src\llm.hpp:1146 - enable llm vision
[DEBUG] src\llm.hpp:1148 - llama.cpp style vision weight
[DEBUG] src\qwen_image.hpp:512  - qwen_image_params.num_layers: 60
[INFO ] src\qwen_image.hpp:514  - use zero_cond_t
[INFO ] src\ggml_extend.hpp:1994 - qwen2.5vl params backend buffer size =  6634.37 MB(VRAM) (857 tensors)
[DEBUG] src\ggml_extend.hpp:1994 - qwen_image params backend buffer size =  12631.04 MB(VRAM) (1933 tensors)
[DEBUG] src\ggml_extend.hpp:1994 - wan_vae params backend buffer size =  242.10 MB(VRAM) (194 tensors)
[DEBUG] stable-diffusion.cpp:752  - Using flash attention in the diffusion model
[INFO ] stable-diffusion.cpp:781  - loading weights
[DEBUG] model.cpp:1350 - using 6 threads for model loading
[DEBUG] model.cpp:1372 - loading tensors from F:\ai\models-image\qwen-image-edit-2511-Q4_K_M.gguf
[DEBUG] 
  |>                                                 | 42/2985 - 204.88it/s�[K
  |=>                                                | 86/2985 - 211.30it/s�[K
  |==>                                               | 129/2985 - 211.48it/s�[K
  |==>                                               | 170/2985 - 209.62it/s�[K
  |===>                                              | 206/2985 - 203.36it/s�[K
  |====>                                             | 259/2985 - 212.99it/s�[K
  |====>                                             | 298/2985 - 210.01it/s�[K
  |=====>                                            | 332/2985 - 204.69it/s�[K
  |======>                                           | 374/2985 - 205.16it/s�[K
  |=======>                                          | 424/2985 - 209.38it/s�[K
  |=======>                                          | 460/2985 - 206.65it/s�[K
  |========>                                         | 504/2985 - 207.66it/s�[K
  |=========>                                        | 552/2985 - 209.97it/s�[K
  |=========>                                        | 588/2985 - 207.63it/s�[K
  |==========>                                       | 639/2985 - 210.61it/s�[K
  |===========>                                      | 712/2985 - 220.02it/s�[K
  |=============>                                    | 778/2985 - 226.36it/s�[K
  |==============>                                   | 844/2985 - 231.93it/s�[K
  |===============>                                  | 912/2985 - 237.44it/s�[K
  |================>                                 | 974/2985 - 240.91it/s�[K
  |=================>                                | 1058/2985 - 249.29it/s�[K
  |==================>                               | 1128/2985 - 253.71it/s�[K
  |====================>                             | 1194/2985 - 256.83it/s�[K
  |=====================>                            | 1258/2985 - 259.38it/s�[K
  |======================>                           | 1322/2985 - 261.73it/s�[K
  |=======================>                          | 1398/2985 - 266.08it/s�[K
  |========================>                         | 1463/2985 - 268.15it/s�[K
  |=========================>                        | 1516/2985 - 267.99it/s�[K
  |==========================>                       | 1578/2985 - 269.28it/s�[K
  |===========================>                      | 1644/2985 - 271.24it/s�[K
  |============================>                     | 1712/2985 - 273.31it/s�[K
  |=============================>                    | 1785/2985 - 276.10it/s�[K
  |==============================>                   | 1849/2985 - 277.34it/s�[K
  |================================>                 | 1922/2985 - 279.77it/s�[K
  |================================>                 | 1934/2985 - 273.55it/s�[Kmodel.cpp:1372 - loading tensors from F:\ai\models-image\Qwen2.5-VL-7B-Instruct.Q4_K_M.gguf

[DEBUG] 
  |================================>                 | 1934/2985 - 273.55it/s�[K
  |=================================>                | 1972/2985 - 271.14it/s�[K
  |=================================>                | 2012/2985 - 269.13it/s�[K
  |==================================>               | 2042/2985 - 265.99it/s�[K
  |==================================>               | 2078/2985 - 263.71it/s�[K
  |===================================>              | 2102/2985 - 260.08it/s�[K
  |===================================>              | 2137/2985 - 257.97it/s�[K
  |====================================>             | 2166/2985 - 255.27it/s�[K
  |====================================>             | 2198/2985 - 252.99it/s�[K
  |=====================================>            | 2234/2985 - 251.24it/s�[K
  |=====================================>            | 2266/2985 - 249.18it/s�[K
  |======================================>           | 2272/2985 - 243.93it/s�[Kmodel.cpp:1372 - loading tensors from F:\ai\models-image\Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf

[DEBUG] 
  |======================================>           | 2272/2985 - 243.93it/s�[K
  |==========================================>       | 2553/2985 - 268.29it/s�[K
  |==============================================>   | 2791/2985 - 286.93it/s�[Kmodel.cpp:1372 - loading tensors from F:\ai\models-image\qwen_image_vae.safetensors

[DEBUG] 
  |==============================================>   | 2791/2985 - 286.93it/s�[K
  |==================================================| 2985/2985 - 299.40it/s�[K
model.cpp:1598 - loading tensors completed, taking 9.98s (process: 0.00s, read: 8.78s, memcpy: 0.00s, convert: 0.10s, copy_to_backend: 0.50s)
[INFO ] stable-diffusion.cpp:816  - finished loaded file
[DEBUG] stable-diffusion.cpp:889  - total params memory size = 19507.51MB (VRAM 19507.51MB, RAM 0.00MB): text_encoders 6634.37MB(VRAM), diffusion_model 12631.05MB(VRAM), vae 242.10MB(VRAM), controlnet 0.00MB(VRAM), pmid 0.00MB(VRAM)
[INFO ] stable-diffusion.cpp:955  - running in FLOW mode
[INFO ] main.cpp:1222 - listening on: 127.0.0.1:5803
[INFO ]