ddvinh1/image-editor-model 🔢🖼️📝 → 🖼️

▶️ 39 runs 📅 Sep 2025 ⚙️ Cog 0.16.3
image-editing image-to-image

About

Example Output

Output

Example output

Performance Metrics

43.80s Prediction Time
244.98s Total Time
All Input Parameters 
{
  "seed": 3,
  "image": "https://replicate.delivery/pbxt/NghtYPUAcWrnShqLZTMbuwX6d9WQklR1kE0o2qoer7faFtne/test_1.jpg",
  "instruction": "replace all the bullets with shimmering, multi-colored butterflies.",
  "guidance_scale": 3,
  "negative_prompt": "low resolution, blur",
  "num_inference_steps": 28,
  "image_guidance_scale": 1.5
}
Input Parameters
seed Type: integerDefault: 3
Random seed
image (required) Type: string
Input image to edit
instruction (required) Type: string
Editing instruction
guidance_scale Type: numberDefault: 3Range: 1 - 10
Text guidance strength (1-10)
negative_prompt Type: stringDefault: low resolution, blur
Negative prompt
num_inference_steps Type: integerDefault: 28Range: 10 - 50
Denoising steps (10-50)
image_guidance_scale Type: numberDefault: 1.5Range: 1 - 10
Image preservation strength (1-10)
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
LlamaModel is using LlamaSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to the manual attention implementation, but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.
2025-09-10 10:05:56,399 - pipeline_hidream_image_editing - INFO - loading editing keys
2025-09-10 10:06:00,404 - pipeline_hidream_image_editing - INFO - finished loading editing keys
  0%|          | 0/28 [00:00<?, ?it/s]
  4%|▎         | 1/28 [00:01<00:36,  1.35s/it]
  7%|▋         | 2/28 [00:02<00:34,  1.33s/it]
 11%|█         | 3/28 [00:03<00:33,  1.33s/it]
 14%|█▍        | 4/28 [00:05<00:31,  1.33s/it]
 18%|█▊        | 5/28 [00:06<00:30,  1.32s/it]
 21%|██▏       | 6/28 [00:07<00:29,  1.32s/it]
 25%|██▌       | 7/28 [00:09<00:27,  1.32s/it]
 29%|██▊       | 8/28 [00:10<00:26,  1.32s/it]
 32%|███▏      | 9/28 [00:11<00:25,  1.32s/it]
 36%|███▌      | 10/28 [00:13<00:23,  1.32s/it]
 39%|███▉      | 11/28 [00:14<00:22,  1.32s/it]
 43%|████▎     | 12/28 [00:15<00:21,  1.32s/it]
 46%|████▋     | 13/28 [00:17<00:19,  1.32s/it]
 50%|█████     | 14/28 [00:18<00:18,  1.32s/it]
 54%|█████▎    | 15/28 [00:19<00:17,  1.32s/it]
 57%|█████▋    | 16/28 [00:21<00:15,  1.32s/it]
 61%|██████    | 17/28 [00:22<00:14,  1.32s/it]
 64%|██████▍   | 18/28 [00:23<00:13,  1.32s/it]
 68%|██████▊   | 19/28 [00:25<00:11,  1.32s/it]2025-09-10 10:06:25,545 - pipeline_hidream_image_editing - INFO - loading refine keys
2025-09-10 10:06:31,167 - pipeline_hidream_image_editing - INFO - finished loading refine keys
2025-09-10 10:06:31,168 - pipeline_hidream_image_editing - INFO - Refining start at step 19
 71%|███████▏  | 20/28 [00:31<00:22,  2.87s/it]
 75%|███████▌  | 21/28 [00:32<00:15,  2.28s/it]
 79%|███████▊  | 22/28 [00:33<00:11,  1.86s/it]
 82%|████████▏ | 23/28 [00:34<00:07,  1.57s/it]
 86%|████████▌ | 24/28 [00:35<00:05,  1.36s/it]
 89%|████████▉ | 25/28 [00:36<00:03,  1.22s/it]
 93%|█████████▎| 26/28 [00:36<00:02,  1.12s/it]
 96%|█████████▋| 27/28 [00:37<00:01,  1.05s/it]
100%|██████████| 28/28 [00:38<00:00,  1.00it/s]
100%|██████████| 28/28 [00:38<00:00,  1.38s/it]
Version Details
Version ID
001795b186ac034be3b22363c23d96c3005009b5d055cb3f49e342827b4c72a5
Version Created
September 11, 2025
Run on Replicate →