zsxkib/infinite-you 🔢📝🖼️❓✓ → 🖼️
About
Transform your portrait photos into any style or setting while preserving your facial identity
Example Output
Prompt:
"a woman with perfect eyes"
Output
Performance Metrics
38.00s
Prediction Time
386.48s
Total Time
All Input Parameters
{
"width": 864,
"height": 1152,
"prompt": "a woman with perfect eyes",
"id_image": "https://replicate.delivery/pbxt/L0gy7uyLE5UP0uz12cndDdSOIgw5R3rV5N6G2pbt7kEK9dCr/0_3.webp",
"num_steps": 30,
"model_version": "sim_stage1",
"enable_realism": true,
"guidance_scale": 3.5,
"enable_anti_blur": true,
"infusenet_guidance_end": 1,
"infusenet_guidance_start": 0,
"infusenet_conditioning_scale": 1
}
Input Parameters
- seed
- Random seed for reproducible results (None generates a random seed)
- width
- Output image width in pixels (recommended: 768, 864, or 960)
- height
- Output image height in pixels (recommended: 960, 1152, or 1280)
- prompt
- Describe how you want the generated image to look. Be specific about details, style, background, etc.
- id_image (required)
- Upload a portrait image containing a human face. For multiple faces, only the largest face will be detected.
- num_steps
- Number of diffusion steps - higher values (30-50) give better quality but take longer
- control_image
- Optional: Upload a second image to control the pose/position of the face in the output
- model_version
- Choose the model version - 'aes_stage2' for better text-image alignment and aesthetics, 'sim_stage1' for higher identity similarity
- output_format
- Choose the format of the output image
- enable_realism
- Apply the realism enhancement LoRA for more realistic-looking results
- guidance_scale
- How closely to follow the prompt (higher = more prompt adherence, lower = more freedom)
- output_quality
- Set the quality of the output image for jpg and webp (1-100)
- enable_anti_blur
- Apply the anti-blur LoRA to reduce blurriness in the results
- infusenet_guidance_end
- Advanced: When to stop applying identity guidance (usually keep at 1.0)
- infusenet_guidance_start
- Advanced: When to start applying identity guidance (0.0-0.1 recommended)
- infusenet_conditioning_scale
- Advanced: Controls how strongly the identity image affects generation (lower values = less identity preservation)
Output Schema
Output
Example Execution Logs
Switching model to sim_stage1 loading lora state dict of realism loading lora state dict of anti_blur Using seed: 3144382269 Preparing ID embeddings Preparing the control image Generating image 0%| | 0/30 [00:00<?, ?it/s] 3%|▎ | 1/30 [00:00<00:21, 1.35it/s] 7%|▋ | 2/30 [00:01<00:18, 1.53it/s] 10%|█ | 3/30 [00:02<00:18, 1.45it/s] 13%|█▎ | 4/30 [00:02<00:18, 1.42it/s] 17%|█▋ | 5/30 [00:03<00:17, 1.40it/s] 20%|██ | 6/30 [00:04<00:17, 1.39it/s] 23%|██▎ | 7/30 [00:04<00:16, 1.39it/s] 27%|██▋ | 8/30 [00:05<00:15, 1.38it/s] 30%|███ | 9/30 [00:06<00:15, 1.38it/s] 33%|███▎ | 10/30 [00:07<00:14, 1.38it/s] 37%|███▋ | 11/30 [00:07<00:13, 1.37it/s] 40%|████ | 12/30 [00:08<00:13, 1.37it/s] 43%|████▎ | 13/30 [00:09<00:12, 1.37it/s] 47%|████▋ | 14/30 [00:10<00:11, 1.37it/s] 50%|█████ | 15/30 [00:10<00:10, 1.37it/s] 53%|█████▎ | 16/30 [00:11<00:10, 1.37it/s] 57%|█████▋ | 17/30 [00:12<00:09, 1.36it/s] 60%|██████ | 18/30 [00:13<00:08, 1.36it/s] 63%|██████▎ | 19/30 [00:13<00:08, 1.37it/s] 67%|██████▋ | 20/30 [00:14<00:07, 1.37it/s] 70%|███████ | 21/30 [00:15<00:06, 1.37it/s] 73%|███████▎ | 22/30 [00:15<00:05, 1.36it/s] 77%|███████▋ | 23/30 [00:16<00:05, 1.36it/s] 80%|████████ | 24/30 [00:17<00:04, 1.36it/s] 83%|████████▎ | 25/30 [00:18<00:03, 1.36it/s] 87%|████████▋ | 26/30 [00:18<00:02, 1.36it/s] 90%|█████████ | 27/30 [00:19<00:02, 1.36it/s] 93%|█████████▎| 28/30 [00:20<00:01, 1.36it/s] 97%|█████████▋| 29/30 [00:21<00:00, 1.36it/s] 100%|██████████| 30/30 [00:21<00:00, 1.36it/s] 100%|██████████| 30/30 [00:21<00:00, 1.37it/s] Saving as WEBP with quality 80
Version Details
- Version ID
b1370c5f5b1bb078eaa87332641c9cc6b89fff1bbd5c61f9e0e81370541b24f0- Version Created
- April 3, 2025