Kvikontent/kviimager2.0

DALL-E 3 XL

Prompt

reimagine the ZX Spectrum Game MANIC MINER as a 3D modern style game

Prompt

cute Harry Potter, pixar animated movie style, dramatic lighting, standing outside Hogwarts.

Prompt

close up of a Quokka, national geographic style photography, stunning image, golden hour

Model description

This is a test model like Dall-E 3.

Estimated generateion time is ~ 60 seconds on gpu

By KVI Kontent

Usage

You can try out model using Huggingface Interface API, and here is how:

import requests
import io
from PIL import *

API_URL = "/static-proxy?url=https%3A%2F%2Fapi-inference.huggingface.co%2Fmodels%2FKvikontent%2Fkviimager2.0"
headers = {"Authorization": "Bearer huggingface_api_token"}

def query(payload):
    response = requests.post(API_URL, headers=headers, json=payload)
    return response.content

image_bytes = query({
    "inputs": "Astronaut riding a horse",
})

image = Image.open(io.BytesIO(image_bytes))
image.save("generated_image.jpg")

or using Diffusers library (requires pytorch and transformers too):

from diffusers import DiffusionPipeline
import io
from PIL import *

pipeline = DiffusionPipeline.from_pretrained("stablediffusionapi/juggernaut-xl-v5")
pipeline.load_lora_weights("Kvikontent/kviimager2.0")

prompt = "Astronaut riding a horse"

image_bytes = pipeline(prompt)
image = Image.open(io.BytesIO(image_bytes))
image.save("generated_image.jpg")

Credits

• Author - Vasiliy Katsyka
• Company - KVIAI
• Licence - Openrail

Official demo

You can use official demo on Spaces: try.