Skip to main content
LiteLLM is a popular Python library that provides a unified interface to 100+ LLM providers. You can use LiteLLM with Lumenfall to access multiple image generation models through a single API.

​
How it works

LiteLLM uses provider prefixes to route requests to different backends. When you use the openai/ prefix with a custom api_base, LiteLLM:
  1. Routes to OpenAI-compatible endpoints: The openai/ prefix tells LiteLLM to use the OpenAI client
  2. Passes the model name through: After stripping the prefix, the model name (e.g., gemini-3-pro-image) is sent directly to your api_base
  3. Preserves parameters: Parameters like size, quality, and style are passed through unchanged
Lumenfall handles parameter transformation on the backend. When you request size="1024x1024" for a model that uses different dimensions, Lumenfall automatically maps it to the closest supported size.

​
Installation

pip install litellm

​
Configuration

Configure LiteLLM to use Lumenfall as an OpenAI-compatible provider: 
import litellm
import os

# Set your Lumenfall API key
os.environ["OPENAI_API_KEY"] = "lmnfl_your_api_key"

# Configure the base URL for Lumenfall
os.environ["OPENAI_API_BASE"] = "https://api.lumenfall.ai/openai/v1"
Alternatively, pass the configuration directly: 
import litellm

litellm.api_key = "lmnfl_your_api_key"
litellm.api_base = "https://api.lumenfall.ai/openai/v1"

​
Generate images

Use litellm.image_generation() to create images: 
import litellm

response = litellm.image_generation(
    model="openai/gemini-3-pro-image",  # prefix with "openai/" for OpenAI-compatible endpoints
    prompt="A serene mountain landscape at sunset with dramatic clouds",
    n=1,
    size="1024x1024",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

image_url = response.data[0].url
print(image_url)

​
Using different models

# Use GPT Image 1.5
response = litellm.image_generation(
    model="openai/gpt-image-1.5",
    prompt="A cyberpunk cityscape",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

# Use Flux.2 Max
response = litellm.image_generation(
    model="openai/flux.2-max",
    prompt="An abstract painting of emotions",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

​
Generation options

response = litellm.image_generation(
    model="openai/gpt-image-1.5",
    prompt="A beautiful garden with roses",
    n=1,
    size="1792x1024",  # landscape orientation
    quality="hd",
    style="natural",
    response_format="url",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

​
Edit images

Use litellm.image_edit() to edit images: 
import litellm

response = litellm.image_edit(
    model="openai/gpt-image-1.5",  # prefix with "openai/" for OpenAI-compatible endpoints
    image=open("original.png", "rb"),
    prompt="Add a rainbow in the sky",
    n=1,
    size="1024x1024",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

edited_url = response.data[0].url
print(edited_url)

​
With a mask

response = litellm.image_edit(
    model="openai/gpt-image-1.5",
    image=open("original.png", "rb"),
    mask=open("mask.png", "rb"),  # transparent areas will be edited
    prompt="Replace with a sunny beach",
    api_key="lmnfl_your_api_key",
    api_base="https://api.lumenfall.ai/openai/v1"
)

​
Async image editing

import asyncio
import litellm

async def edit_image():
    response = await litellm.aimage_edit(
        model="openai/gpt-image-1.5",
        image=open("original.png", "rb"),
        prompt="Make it look like winter",
        api_key="lmnfl_your_api_key",
        api_base="https://api.lumenfall.ai/openai/v1"
    )
    return response.data[0].url

url = asyncio.run(edit_image())
print(url)

​
LiteLLM Proxy

You can also use the LiteLLM Proxy server to route requests to Lumenfall. Add this to your litellm_config.yaml: 
model_list:
  - model_name: lumenfall-gemini-image
    litellm_params:
      model: openai/gemini-3-pro-image
      api_key: lmnfl_your_api_key
      api_base: https://api.lumenfall.ai/openai/v1

  - model_name: lumenfall-gpt-image
    litellm_params:
      model: openai/gpt-image-1.5
      api_key: lmnfl_your_api_key
      api_base: https://api.lumenfall.ai/openai/v1

  - model_name: lumenfall-flux
    litellm_params:
      model: openai/flux.2-max
      api_key: lmnfl_your_api_key
      api_base: https://api.lumenfall.ai/openai/v1
Start the proxy: 
litellm --config litellm_config.yaml
Then make requests to your proxy: 
import litellm

response = litellm.image_generation(
    model="lumenfall-gemini-image",
    prompt="A serene lake at dawn",
    api_base="http://localhost:4000"  # LiteLLM proxy URL
)

​
Error handling

from litellm import APIError, AuthenticationError, RateLimitError

try:
    response = litellm.image_generation(
        model="openai/gemini-3-pro-image",
        prompt="A beautiful sunset",
        api_key="lmnfl_your_api_key",
        api_base="https://api.lumenfall.ai/openai/v1"
    )
except AuthenticationError:
    print("Invalid API key")
except RateLimitError:
    print("Rate limit exceeded")
except APIError as e:
    print(f"API error: {e}")

​
Async support

import asyncio
import litellm

async def generate_image():
    response = await litellm.aimage_generation(
        model="openai/gemini-3-pro-image",
        prompt="A futuristic space station",
        api_key="lmnfl_your_api_key",
        api_base="https://api.lumenfall.ai/openai/v1"
    )
    return response.data[0].url

# Run async function
url = asyncio.run(generate_image())
print(url)

​
Next steps

API Reference

Explore the full API documentation.

Available Models

See all available image generation models.