fwus_ai/geely-car2go 🖼️🔢❓📝✓ → 🖼️

▶️ 78 runs 📅 Aug 2025 ⚙️ Cog 0.13.6

About

Example Output

Prompt:

"A sleek, metallic blue Geely Coolray TOK speeds through a dimly lit tunnel, captured from a low, frontal perspective that emphasizes its aggressive stance and bold, modern design. The vehicle’s distinctive grille and sharp LED headlights cut through the warm amber tones of the tunnel walls, radiating high-tech confidence. An orange round logo with bold white text reading “car2go” is prominently displayed on both front doors and the center of the hood, standing out vividly against the blue paint. Overhead tunnel lights create cinematic lens flares and subtle reflections across the polished bodywork, enhancing the sense of motion and drama. The tunnel’s ribbed wall panels stretch into the distance toward a bright daylight opening, reinforcing the vehicle’s speed and forward momentum. Slight motion blur trails behind the moving car, while its aerodynamic contours gleam under the low artificial lighting, blending deep shadows with luminous highlights. The orange “car2go” logos catch stray light, adding a pop of vibrant contrast. The centered, symmetrical camera framing positions the Geely Coolray TOK as the undeniable focal point, exuding dominance and modern urban style. Ultra-detailed, high-contrast photorealism, cinematic lighting, motion blur, and precise reflections give the image a premium automotive commercial feel."

Output

Performance Metrics

40.35s Prediction Time

40.72s Total Time

All Input Parameters

{
  "model": "dev",
  "prompt": "A sleek, metallic blue Geely Coolray TOK speeds through a dimly lit tunnel, captured from a low, frontal perspective that emphasizes its aggressive stance and bold, modern design. The vehicle’s distinctive grille and sharp LED headlights cut through the warm amber tones of the tunnel walls, radiating high-tech confidence. An orange round logo with bold white text reading “car2go” is prominently displayed on both front doors and the center of the hood, standing out vividly against the blue paint. Overhead tunnel lights create cinematic lens flares and subtle reflections across the polished bodywork, enhancing the sense of motion and drama. The tunnel’s ribbed wall panels stretch into the distance toward a bright daylight opening, reinforcing the vehicle’s speed and forward momentum. Slight motion blur trails behind the moving car, while its aerodynamic contours gleam under the low artificial lighting, blending deep shadows with luminous highlights. The orange “car2go” logos catch stray light, adding a pop of vibrant contrast. The centered, symmetrical camera framing positions the Geely Coolray TOK as the undeniable focal point, exuding dominance and modern urban style. Ultra-detailed, high-contrast photorealism, cinematic lighting, motion blur, and precise reflections give the image a premium automotive commercial feel.\n",
  "go_fast": false,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 3,
  "aspect_ratio": "16:9",
  "output_format": "png",
  "guidance_scale": 3.29,
  "output_quality": 80,
  "prompt_strength": 0.8,
  "extra_lora_scale": 1,
  "num_inference_steps": 48
}

Input Parameters

mask Type: string: Image mask for image inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
seed Type: integer: Random seed. Set for reproducible generation
image Type: string: Input image for image to image or inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
model Default: dev: Which model to run inference with. The dev model performs best with around 28 inference steps but the schnell model only needs 4 steps.
width Type: integerRange: 256 - 1440: Width of generated image. Only works if `aspect_ratio` is set to custom. Will be rounded to nearest multiple of 16. Incompatible with fast generation
height Type: integerRange: 256 - 1440: Height of generated image. Only works if `aspect_ratio` is set to custom. Will be rounded to nearest multiple of 16. Incompatible with fast generation
prompt (required) Type: string: Prompt for generated image. If you include the `trigger_word` used in the training process you are more likely to activate the trained object, style, or concept in the resulting image.
go_fast Type: booleanDefault: false: Run faster predictions with model optimized for speed (currently fp8 quantized); disable to run in original bf16
extra_lora Type: string: Load LoRA weights. Supports Replicate models in the format <owner>/<username> or <owner>/<username>/<version>, HuggingFace URLs in the format huggingface.co/<owner>/<model-name>, CivitAI URLs in the format civitai.com/models/<id>[/<model-name>], or arbitrary .safetensors URLs from the Internet. For example, 'fofr/flux-pixar-cars'
lora_scale Type: numberDefault: 1Range: -1 - 3: Determines how strongly the main LoRA should be applied. Sane results between 0 and 1 for base inference. For go_fast we apply a 1.5x multiplier to this value; we've generally seen good performance when scaling the base value by that amount. You may still need to experiment to find the best value for your particular lora.
megapixels Default: 1: Approximate number of megapixels for generated image
num_outputs Type: integerDefault: 1Range: 1 - 4: Number of outputs to generate
aspect_ratio Default: 1:1: Aspect ratio for the generated image. If custom is selected, uses height and width below & will run in bf16 mode
output_format Default: webp: Format of the output images
guidance_scale Type: numberDefault: 3Range: 0 - 10: Guidance scale for the diffusion process. Lower values can give more realistic images. Good values to try are 2, 2.5, 3 and 3.5
output_quality Type: integerDefault: 80Range: 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
prompt_strength Type: numberDefault: 0.8Range: 0 - 1: Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
extra_lora_scale Type: numberDefault: 1Range: -1 - 3: Determines how strongly the extra LoRA should be applied. Sane results between 0 and 1 for base inference. For go_fast we apply a 1.5x multiplier to this value; we've generally seen good performance when scaling the base value by that amount. You may still need to experiment to find the best value for your particular lora.
replicate_weights Type: string: Load LoRA weights. Supports Replicate models in the format <owner>/<username> or <owner>/<username>/<version>, HuggingFace URLs in the format huggingface.co/<owner>/<model-name>, CivitAI URLs in the format civitai.com/models/<id>[/<model-name>], or arbitrary .safetensors URLs from the Internet. For example, 'fofr/flux-pixar-cars'
num_inference_steps Type: integerDefault: 28Range: 1 - 50: Number of denoising steps. More steps can give more detailed images, but take longer.
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

free=25367092572160
Downloading weights
2025-08-11T13:07:46Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmp0ah5ef1y/weights url=https://replicate.delivery/xezq/BV1QAZDdXIYWLVdkGPx60Deevmez36HsfBrIdjoLBGbseNPpC/flux-lora.tar
2025-08-11T13:07:46Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local
2025-08-11T13:07:46Z | INFO  | [ Cache URL Rewrite ] enabled=true target_url=http://hermes.services.svc.cluster.local/replicate.delivery/xezq/BV1QAZDdXIYWLVdkGPx60Deevmez36HsfBrIdjoLBGbseNPpC/flux-lora.tar url=https://replicate.delivery/xezq/BV1QAZDdXIYWLVdkGPx60Deevmez36HsfBrIdjoLBGbseNPpC/flux-lora.tar
2025-08-11T13:07:46Z | INFO  | [ Redirect ] redirect_url=http://r8-east4-loras-ric1.cwlota.com/c704fac74582fe7b3d60c3a52ee009d53ed527f95845173f8a7d5aa4858946e9?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Checksum-Mode=ENABLED&X-Amz-Credential=CWNZUVKLDHXVHEZN%2F20250811%2FUS-EAST-04A%2Fs3%2Faws4_request&X-Amz-Date=20250811T130746Z&X-Amz-Expires=600&X-Amz-SignedHeaders=host&x-id=GetObject&X-Amz-Signature=f7616f022fed5729f383fd002766c43a1fdc3520f17ba200ce11252f36073af0 url=http://hermes.services.svc.cluster.local/replicate.delivery/xezq/BV1QAZDdXIYWLVdkGPx60Deevmez36HsfBrIdjoLBGbseNPpC/flux-lora.tar
2025-08-11T13:07:46Z | INFO  | [ Complete ] dest=/tmp/tmp0ah5ef1y/weights size="172 MB" total_elapsed=0.619s url=https://replicate.delivery/xezq/BV1QAZDdXIYWLVdkGPx60Deevmez36HsfBrIdjoLBGbseNPpC/flux-lora.tar
Downloaded weights in 0.68s
Loaded LoRAs in 1.64s
Using seed: 11695
Prompt: A sleek, metallic blue Geely Coolray TOK speeds through a dimly lit tunnel, captured from a low, frontal perspective that emphasizes its aggressive stance and bold, modern design. The vehicle’s distinctive grille and sharp LED headlights cut through the warm amber tones of the tunnel walls, radiating high-tech confidence. An orange round logo with bold white text reading “car2go” is prominently displayed on both front doors and the center of the hood, standing out vividly against the blue paint. Overhead tunnel lights create cinematic lens flares and subtle reflections across the polished bodywork, enhancing the sense of motion and drama. The tunnel’s ribbed wall panels stretch into the distance toward a bright daylight opening, reinforcing the vehicle’s speed and forward momentum. Slight motion blur trails behind the moving car, while its aerodynamic contours gleam under the low artificial lighting, blending deep shadows with luminous highlights. The orange “car2go” logos catch stray light, adding a pop of vibrant contrast. The centered, symmetrical camera framing positions the Geely Coolray TOK as the undeniable focal point, exuding dominance and modern urban style. Ultra-detailed, high-contrast photorealism, cinematic lighting, motion blur, and precise reflections give the image a premium automotive commercial feel.
[!] txt2img mode
  0%|          | 0/48 [00:00<?, ?it/s]
  2%|▏         | 1/48 [00:00<00:33,  1.38it/s]
  4%|▍         | 2/48 [00:01<00:29,  1.56it/s]
  6%|▋         | 3/48 [00:02<00:30,  1.47it/s]
  8%|▊         | 4/48 [00:02<00:30,  1.43it/s]
 10%|█         | 5/48 [00:03<00:30,  1.41it/s]
 12%|█▎        | 6/48 [00:04<00:30,  1.40it/s]
 15%|█▍        | 7/48 [00:04<00:29,  1.39it/s]
 17%|█▋        | 8/48 [00:05<00:28,  1.39it/s]
 19%|█▉        | 9/48 [00:06<00:28,  1.38it/s]
 21%|██        | 10/48 [00:07<00:27,  1.38it/s]
 23%|██▎       | 11/48 [00:07<00:26,  1.38it/s]
 25%|██▌       | 12/48 [00:08<00:26,  1.38it/s]
 27%|██▋       | 13/48 [00:09<00:25,  1.38it/s]
 29%|██▉       | 14/48 [00:10<00:24,  1.38it/s]
 31%|███▏      | 15/48 [00:10<00:23,  1.38it/s]
 33%|███▎      | 16/48 [00:11<00:23,  1.38it/s]
 35%|███▌      | 17/48 [00:12<00:22,  1.38it/s]
 38%|███▊      | 18/48 [00:12<00:21,  1.38it/s]
 40%|███▉      | 19/48 [00:13<00:21,  1.38it/s]
 42%|████▏     | 20/48 [00:14<00:20,  1.38it/s]
 44%|████▍     | 21/48 [00:15<00:19,  1.38it/s]
 46%|████▌     | 22/48 [00:15<00:18,  1.38it/s]
 48%|████▊     | 23/48 [00:16<00:18,  1.38it/s]
 50%|█████     | 24/48 [00:17<00:17,  1.38it/s]
 52%|█████▏    | 25/48 [00:17<00:16,  1.38it/s]
 54%|█████▍    | 26/48 [00:18<00:15,  1.38it/s]
 56%|█████▋    | 27/48 [00:19<00:15,  1.38it/s]
 58%|█████▊    | 28/48 [00:20<00:14,  1.38it/s]
 60%|██████    | 29/48 [00:20<00:13,  1.38it/s]
 62%|██████▎   | 30/48 [00:21<00:13,  1.38it/s]
 65%|██████▍   | 31/48 [00:22<00:12,  1.38it/s]
 67%|██████▋   | 32/48 [00:23<00:11,  1.38it/s]
 69%|██████▉   | 33/48 [00:23<00:10,  1.38it/s]
 71%|███████   | 34/48 [00:24<00:10,  1.38it/s]
 73%|███████▎  | 35/48 [00:25<00:09,  1.38it/s]
 75%|███████▌  | 36/48 [00:25<00:08,  1.38it/s]
 77%|███████▋  | 37/48 [00:26<00:07,  1.38it/s]
 79%|███████▉  | 38/48 [00:27<00:07,  1.38it/s]
 81%|████████▏ | 39/48 [00:28<00:06,  1.38it/s]
 83%|████████▎ | 40/48 [00:28<00:05,  1.38it/s]
 85%|████████▌ | 41/48 [00:29<00:05,  1.38it/s]
 88%|████████▊ | 42/48 [00:30<00:04,  1.38it/s]
 90%|████████▉ | 43/48 [00:31<00:03,  1.38it/s]
 92%|█████████▏| 44/48 [00:31<00:02,  1.38it/s]
 94%|█████████▍| 45/48 [00:32<00:02,  1.38it/s]
 96%|█████████▌| 46/48 [00:33<00:01,  1.38it/s]
 98%|█████████▊| 47/48 [00:33<00:00,  1.38it/s]
100%|██████████| 48/48 [00:34<00:00,  1.38it/s]
100%|██████████| 48/48 [00:34<00:00,  1.39it/s]
Total safe images: 3 out of 3

Version Details

Version ID: b89216a0c14ea29fb9ba777deb1e4b01c9b57493aecab8010cb78aab5e156858
Version Created: August 11, 2025

Run on Replicate →