pollinations.ai - Image Generation

pollinations.ai: (https://pollinations.ai/)

Work with the best generative models from Pollinations using this python wrapper.

Installing

pip install -U pollinations
pip install -U pollinations.ai

# Linux/macOS
python3 -m pip install -U pollinations
python3 -m pip install -U pollinations.ai

# Windows
py -3 -m pip install -U pollinations
py -3 -m pip install -U pollinations.ai

Image Model

import pollinations

image_model: pollinations.ImageModel = pollinations.image(
    model = pollinations.image_default,
    seed = 0,
    width = 1024,
    height = 1024,
    enhance = False,
    nologo = False,
    private = False,
)

image_model.generate(
    prompt = "A black cat in a cyberpunk city.",
    negative = "Anime, cartoony, childish.",
    save = True,
    file = "image-output.png",
)

Text Model

import pollinations

text_model: pollinations.TextModel = pollinations.text(
    frequency_penalty = 0,
    presence_penalty = 0,
    temperature = 0.5,
    top_p = 1,
    model = pollinations.text_default,
    stream = True,
    contextual = True, # True: Holds conversation context up to 10. False: Has no conversation context
    system = "You are a polite AI Assistant named Pollinations! Use emojis and markdown as you wish."
)

text_model.generate(
    prompt="What is 1+1?",
    display=True
)
text_model.generate(
    prompt="Now add 10 to that.",
    display=True
)

Multi Model (Image & Text)

import pollinations

multi_model: pollinations.MultiModel = pollinations.multi(
    system = "You are a polite AI Assistant named Pollinations! Use emojis and markdown as you wish.",
    default = None, # None: AI will infer what model to use. Example: pollinations.turbo: Will default image model to turbo
    text_model = pollinations.text_default, # Safety fail measure incase of model errors in pollinations api.
    image_model = pollinations.image_default, # Safety fail measure incase of model errors in pollinations api.
)

multi_model.generate(
    "Hi",
    display=True,
    provide_details=False # Provides the details and objects of each generation
)
multi_model.generate(
    "Make an image of a black dog in a cyberpunk city.",
    display=True,
    provide_details=False
)
multi_model.generate(
    "Thanks.",
    display=True,
    provide_details=False
)

Smart Model (MultiModel up-to-date with time, dates, weather, and search) (Primitive Testing)

import pollinations

# Searching will not work unless you provide a serpapi api-key like this:
pollinations.keys(serpapi="your-key")

smart_model: pollinations.SmartModel = pollinations.smart(
    system="You are a helpful and friendly AI assistant. Use emojis and markdown as you like.",
    text_model=pollinations.mistral_large,      # Optional
    # image_model=pollinations.flux_anime       # Optional : If not directly chosen, the best fit model according to prompt will be chosen.
)

smart_model.generate(
    prompt="Hi.",
    display=True,
    provide_details=False    # Provide extra details of each generation.
)
smart_model.generate(
    prompt="What is the weather in london like?",
    display=True,
    provide_details=False
)
smart_model.generate(
    prompt="What's the latest news there as well?",
    display=True,
    provide_details=False
)
smart_model.generate(
    prompt="What time is it in New York City?",
    display=True,
    provide_details=False
)
smart_model.generate(
    prompt="Make an image of that at night, include city lights.",
    display=True,
    provide_details=False
)
smart_model.generate(
    prompt="Thanks.",
    display=True,
    provide_details=False
)

Links

• Pollinations.ai
• Discord
• Github
• Youtube
• Instagram
• Twitter (X)