qwen/qwen-edit-multiangle 🔢🖼️📝✓❓ → 🖼️

⭐ Official ▶️ 298.7K runs 📅 Nov 2025 ⚙️ Cog 0.16.8 📄 Paper ⚖️ License

camera-aware image-editing image-to-image

About

Camera-aware edits for Qwen/Qwen-Image-Edit-2509 with Lightning + multi-angle LoRA

Example Output

Output

Performance Metrics

64.34s Prediction Time

64.35s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/O1W5PaLjEt69eCt8VsKREQTtoh2t8dBhfT8mU839T0wTfpaw/capy.webp",
  "prompt": "",
  "go_fast": false,
  "aspect_ratio": "match_input_image",
  "move_forward": 0,
  "output_format": "webp",
  "vertical_tilt": 0,
  "output_quality": 95,
  "rotate_degrees": 90,
  "use_wide_angle": false,
  "true_guidance_scale": 1,
  "use_multiple_angles": true,
  "multiple_angles_strength": 1.25
}

Input Parameters

seed Type: integer: Random seed. Set for reproducible generation.
image (required) Type: string: Image file uploaded to Cog (jpeg, png, gif, or webp).
prompt Type: stringDefault:: Optional text instruction appended after the camera directive.
go_fast Type: booleanDefault: true: If num_inference_steps is omitted, true runs the 4-step fast preset and false runs the 40-step detailed preset.
lora_scale Type: numberDefault: 1.25Range: 0 - 4: Strength applied to the selected LoRA.
aspect_ratio Default: match_input_image: Aspect ratio for the generated image
lora_weights Type: stringDefault: dx8152/Qwen-Edit-2509-Multiple-angles: LoRA weights to apply. Pass a Hugging Face repo slug like 'dx8152/Qwen-Edit-2509-Multiple-angles' or a direct .safetensors/zip/tar URL (for example, 'https://huggingface.co/flymy-ai/qwen-image-lora/resolve/main/pytorch_lora_weights.safetensors', 'https://example.com/lora_weights.tar.gz', or 'https://example.com/lora_weights.zip').
move_forward Type: integerDefault: 0Range: 0 - 10: Move the camera forward (zoom in). Higher values push toward a close-up framing.
output_format Default: webp: Format of the output images.
vertical_tilt Type: integerDefault: 0Range: -1 - 1: Vertical camera tilt. -1 = bird's-eye, 0 = level, 1 = worm's-eye.
output_quality Type: integerDefault: 95Range: 0 - 100: Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs.
rotate_degrees Type: integerDefault: 0Range: -90 - 90: Camera rotation in degrees. Positive rotates left, negative rotates right.
use_wide_angle Type: booleanDefault: false: Switch to a wide-angle lens instruction.
num_inference_steps Type: integerRange: 1 - 40: Explicit denoising step count (1-40). Leave blank to use the go_fast presets (4 or 40 steps).
true_guidance_scale Type: numberDefault: 1Range: 0 - 10: True classifier-free guidance scale passed to the pipeline.
disable_safety_checker Type: booleanDefault: false: Disable safety checker for generated images.

Output Schema

Output

Type: array • Items Type: string • Items Format: uri

Example Execution Logs

Using seed: 793851
Active adapters: multiple_angles
Generation took 64.09 seconds
/root/.pyenv/versions/3.11.13/lib/python3.11/site-packages/cog/server/scope.py:22: ExperimentalFeatureWarning: current_scope is an experimental internal function. It may change or be removed without warning.
warnings.warn(
Total safe images: 1 out of 1

Version Details

Version ID: cf245ffaa67a6d7d0edeb597d2fded5ab80cbf72b0dceec185d709ea99667f79
Version Created: November 11, 2025

Run on Replicate →