lucataco/omnigen2 🔢🖼️📝❓ → 🖼️
About
OmniGen2: a powerful and efficient unified multimodal model

Example Output
Prompt:
"Change the dress to blue"
Output

Performance Metrics
93.17s
Prediction Time
200.64s
Total Time
All Input Parameters
{ "seed": -1, "image": "https://replicate.delivery/pbxt/NFVLgKWVv48pYj7p8qCnKpeYDN5S46MlxYNaIEzRu03KG1xJ/yellow-dress.png", "width": 1024, "height": 1024, "prompt": "Change the dress to blue", "scheduler": "euler", "max_pixels": 1048576, "cfg_range_end": 1, "cfg_range_start": 0, "negative_prompt": "(((deformed))), blurry, over saturation, bad anatomy, disfigured, poorly drawn face, mutation, mutated, (extra_limb), (ugly), (poorly drawn hands), fused fingers, messy drawing, broken legs censor, censored, censor_bar", "num_inference_steps": 50, "text_guidance_scale": 5, "image_guidance_scale": 2, "max_input_image_side_length": 2048 }
Input Parameters
- seed
- Random seed. Set to -1 for random seed
- image (required)
- Input image to edit
- width
- Width of output image
- height
- Height of output image
- prompt
- Text prompt describing the desired image edit
- image_2
- Optional second input image for multi-image operations
- image_3
- Optional third input image for multi-image operations
- scheduler
- Scheduler to use
- max_pixels
- Maximum number of pixels in output
- cfg_range_end
- CFG range end
- cfg_range_start
- CFG range start
- negative_prompt
- Negative prompt to guide what should not be in the image
- num_inference_steps
- Number of denoising steps
- text_guidance_scale
- Guidance scale for text prompt
- image_guidance_scale
- Guidance scale for input image. Higher values increase consistency with input image
- max_input_image_side_length
- Maximum input image side length
Output Schema
Output
Example Execution Logs
0%| | 0/50 [00:00<?, ?it/s] 2%|▏ | 1/50 [00:05<04:53, 5.99s/it] 4%|▍ | 2/50 [00:07<02:42, 3.39s/it] 6%|▌ | 3/50 [00:09<02:03, 2.64s/it] 8%|▊ | 4/50 [00:11<01:44, 2.28s/it] 10%|█ | 5/50 [00:12<01:33, 2.08s/it] 12%|█▏ | 6/50 [00:14<01:26, 1.97s/it] 14%|█▍ | 7/50 [00:16<01:21, 1.89s/it] 16%|█▌ | 8/50 [00:17<01:17, 1.84s/it] 18%|█▊ | 9/50 [00:19<01:14, 1.81s/it] 20%|██ | 10/50 [00:21<01:11, 1.79s/it] 22%|██▏ | 11/50 [00:23<01:09, 1.78s/it] 24%|██▍ | 12/50 [00:24<01:07, 1.77s/it] 26%|██▌ | 13/50 [00:26<01:05, 1.76s/it] 28%|██▊ | 14/50 [00:28<01:03, 1.76s/it] 30%|███ | 15/50 [00:30<01:01, 1.76s/it] 32%|███▏ | 16/50 [00:31<00:59, 1.75s/it] 34%|███▍ | 17/50 [00:33<00:57, 1.75s/it] 36%|███▌ | 18/50 [00:35<00:56, 1.75s/it] 38%|███▊ | 19/50 [00:37<00:54, 1.75s/it] 40%|████ | 20/50 [00:38<00:52, 1.75s/it] 42%|████▏ | 21/50 [00:40<00:50, 1.76s/it] 44%|████▍ | 22/50 [00:42<00:49, 1.76s/it] 46%|████▌ | 23/50 [00:44<00:47, 1.76s/it] 48%|████▊ | 24/50 [00:46<00:45, 1.76s/it] 50%|█████ | 25/50 [00:47<00:44, 1.76s/it] 52%|█████▏ | 26/50 [00:49<00:42, 1.76s/it] 54%|█████▍ | 27/50 [00:51<00:40, 1.76s/it] 56%|█████▌ | 28/50 [00:53<00:38, 1.76s/it] 58%|█████▊ | 29/50 [00:54<00:37, 1.76s/it] 60%|██████ | 30/50 [00:56<00:35, 1.76s/it] 62%|██████▏ | 31/50 [00:58<00:33, 1.77s/it] 64%|██████▍ | 32/50 [01:00<00:31, 1.77s/it] 66%|██████▌ | 33/50 [01:01<00:29, 1.76s/it] 68%|██████▊ | 34/50 [01:03<00:28, 1.76s/it] 70%|███████ | 35/50 [01:05<00:26, 1.76s/it] 72%|███████▏ | 36/50 [01:07<00:24, 1.76s/it] 74%|███████▍ | 37/50 [01:08<00:22, 1.76s/it] 76%|███████▌ | 38/50 [01:10<00:21, 1.76s/it] 78%|███████▊ | 39/50 [01:12<00:19, 1.76s/it] 80%|████████ | 40/50 [01:14<00:17, 1.76s/it] 82%|████████▏ | 41/50 [01:16<00:15, 1.76s/it] 84%|████████▍ | 42/50 [01:17<00:14, 1.76s/it] 86%|████████▌ | 43/50 [01:19<00:12, 1.76s/it] 88%|████████▊ | 44/50 [01:21<00:10, 1.76s/it] 90%|█████████ | 45/50 [01:23<00:08, 1.77s/it] 92%|█████████▏| 46/50 [01:24<00:07, 1.77s/it] 94%|█████████▍| 47/50 [01:26<00:05, 1.77s/it] 96%|█████████▌| 48/50 [01:28<00:03, 1.77s/it] 98%|█████████▊| 49/50 [01:30<00:01, 1.77s/it] 100%|██████████| 50/50 [01:31<00:00, 1.77s/it] 100%|██████████| 50/50 [01:31<00:00, 1.84s/it]
Version Details
- Version ID
5b9ea1d0821a60be9c861ebfc3513d121ecd8cab1932d3aa8d703e517988502e
- Version Created
- June 25, 2025