imagepipeline/RunDiffusion-XL

1---2license: creativeml-openrail-m3tags:4  - imagepipeline5  - imagepipeline.io6  - text-to-image7  - ultra-realistic8pinned: false9pipeline_tag: text-to-image10 11---12 13 14## RunDiffusion-XL15<img src="https://f005.backblazeb2.com/b2api/v2/b2_download_file_by_id?fileId=4_zfdf0a8ed59e8666b89b10713_f110328b9d25ecaad_d20231130_m135905_c005_v0501015_t0042_u01701352745784" alt="Generated by Image Pipeline" style="border-radius: 10px;">16 17 18 19**This checkpoint model is uploaded on [imagepipeline.io](https://imagepipeline.io/)**20 21Model details - Strengths22 23More photorealistic for animals, cars, and people.24 25Different Model, different generations. Try it with your favorite prompt!26 27Better portraits. SDXL likes to put the subject way far out. This brings things closer.28 29Eyes and hands are a bit better (from our testing)30 31Better prompt control. Make sure to use commas and break apart subjects and styles!32 33Weaknesses34 35Still a few concepts that are "plastic" (Food, but please test your prompts and let us know!)36 37Can sometimes be too "person bias"38 39Creativity is a little bit hindered. This could be considered a "Photorealistic" model. But again, TEST!40 41Bias towards "sports cars" lol42 43Notice the RDXL follows the prompt a bit better. When asking for "upper body" it tends to follow that.44 45How to Use46 47Prompt as usual. Use commas to separate subjects and styles. If you aren't getting what you want, use the token again. e.g. iron man made of lava, suit of armor infused with molten lava, ...etc if you don't get Iron Man in your generation, then put that token somewhere else, `iron man made of lava, suit of armor infused with molten lava, marvel iron man super hero ...etc`48 49Use the standard refiner or the 1.0-0.9 refiner. 50 51 52 53 54[![Try this model](https://img.shields.io/badge/try_this_model-image_pipeline-BD9319)](https://imagepipeline.io/models/RunDiffusion-XL?id=372f1620-2665-46fa-985d-c68b41d0f6da/) 55 56 57 58 59## How to try this model ?60 61You can try using it locally or send an API call to test the output quality.62 63Get your `API_KEY` from  [imagepipeline.io](https://imagepipeline.io/). No payment required.64 65 66 67 68 69Coding in `php` `javascript` `node` etc ? Checkout our documentation  70 71[![documentation](https://img.shields.io/badge/documentation-image_pipeline-blue)](https://docs.imagepipeline.io/docs/introduction) 72 73 74```python75import requests  76import json  77  78url =  "https://imagepipeline.io/sdxl/text2image/v1/run"  79  80payload = json.dumps({  81"model_id":  "372f1620-2665-46fa-985d-c68b41d0f6da",  82"prompt":  "ultra realistic close up portrait ((beautiful pale cyberpunk female with heavy black eyeliner)), blue eyes, shaved side haircut, hyper detail, cinematic lighting, magic neon, dark red city, Canon EOS R3, nikon, f/1.4, ISO 200, 1/160s, 8K, RAW, unedited, symmetrical balance, in-frame, 8K",  83"negative_prompt":  "painting, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, deformed, ugly, blurry, bad anatomy, bad proportions, extra limbs, cloned face, skinny, glitchy, double torso, extra arms, extra hands, mangled fingers, missing lips, ugly face, distorted face, extra legs, anime",  84"width":  "512",  85"height":  "512",  86"samples":  "1",  87"num_inference_steps":  "30",  88"safety_checker":  false,   89"guidance_scale":  7.5,  90"multi_lingual":  "no",  91"embeddings":  "", 92"lora_models": "", 93"lora_weights":  "" 94})  95  96headers =  {  97'Content-Type':  'application/json',98'API-Key': 'your_api_key'99}  100  101response = requests.request("POST", url, headers=headers, data=payload)  102  103print(response.text)104 105}106```107 108Get more ready to use `MODELS` like this for `SD 1.5` and `SDXL` : 109 110[![All models](https://img.shields.io/badge/Get%20All%20Models-image_pipeline-BD9319)](https://imagepipeline.io/models) 111 112### API Reference113 114#### Generate Image115 116```http117  https://api.imagepipeline.io/sdxl/text2image/v1118```119 120| Headers               | Type     | Description                                                                                                        |121|:----------------------| :------- |:-------------------------------------------------------------------------------------------------------------------|122| `API-Key`             | `str` | Get your `API_KEY` from  [imagepipeline.io](https://imagepipeline.io/)                                             |123| `Content-Type`        | `str` | application/json - content type of the request body |124 125 126| Parameter | Type     | Description                |127| :-------- | :------- | :------------------------- |128| `model_id` | `str` | Your base model, find available lists in  [models page](https://imagepipeline.io/models) or upload your own|129| `prompt` | `str` | Text Prompt. Check our [Prompt Guide](https://docs.imagepipeline.io/docs/SD-1.5/docs/extras/prompt-guide) for tips |130| `num_inference_steps` | `int [1-50]` | Noise is removed with each step, resulting in a higher-quality image over time. Ideal value 30-50 (without LCM) |131| `guidance_scale` | `float [1-20]` | Higher guidance scale prioritizes text prompt relevance but sacrifices image quality. Ideal value 7.5-12.5 |132| `lora_models` | `str, array` | 	Pass the model_id(s) of LoRA models that can be found in models page |133| `lora_weights` | `str, array` | Strength of the LoRA effect |134 135---136license: creativeml-openrail-m137tags:138  - imagepipeline139  - imagepipeline.io140  - text-to-image141  - ultra-realistic142pinned: false143pipeline_tag: text-to-image144 145---146 147### Feedback 148 149If you have any feedback, please reach out to us at hello@imagepipeline.io150 151 152#### 🔗 Visit Website153[![portfolio](https://img.shields.io/badge/image_pipeline-BD9319?style=for-the-badge&logo=gocd&logoColor=white)](https://imagepipeline.io/)154 155 156If you are the original author of this model, please [click here](https://airtable.com/apprTaRnJbDJ8ufOx/shr4g7o9B6fWfOlUR) to add credits157