# OmniGen v2

> Unified multimodal model for text-to-image generation, instruction-guided image editing, personalized generation, and virtual try-on

## Quick Reference

- Model ID: omnigen-v2
- Creator: VectorSpaceLab
- Status: active
- Family: omnigen
- Base URL: https://api.lumenfall.ai/openai/v1

## Specifications

- Max Input Images: 3
- Input Modalities: text, image
- Output Modalities: image
- Supported Modes: Text to Image, Image Edit

## API Parameters

The compiled parameter schema for this model is available via the API: `GET /v1/models/omnigen-v2?schema=true`.

### Core Parameters

- `prompt` (string) — REQUIRED: Text prompt for image generation. Modes: Text to Image, Image Edit
- `negative_prompt` (string): Negative prompt to guide generation away from undesired content. Modes: Text to Image, Image Edit
- `seed` (integer): Random seed for reproducibility. Modes: Text to Image, Image Edit

### Size & Layout

- `size` (string): Image dimensions as WxH pixels (e.g. "1024x1024") or aspect ratio (e.g. "16:9"). Modes: Text to Image, Image Edit
- `aspect_ratio` (string): Aspect ratio of the output image (e.g. "16:9", "1:1"). Modes: Text to Image, Image Edit
- `resolution` (string): Output resolution tier (e.g. "1K", "4K"). Modes: Text to Image, Image Edit

### Media Inputs

- `image` (file) — REQUIRED: Input image(s) to edit. Modes: Image Edit

### Output & Format

- `response_format` (string): How to return the image. Default: url. Values: url, b64_json. Modes: Text to Image, Image Edit
- `output_format` (string): Output image format. Values: png, jpeg, gif, webp, avif. Modes: Text to Image, Image Edit
- `output_compression` (integer): Compression level for lossy formats (JPEG, WebP, AVIF). Modes: Text to Image, Image Edit
- `n` (integer): Number of images to generate. Default: 1. Modes: Text to Image, Image Edit

### Additional Parameters

- `cfg_range_end` (number): CFG range end value.. Modes: Text to Image, Image Edit
- `cfg_range_start` (number): CFG range start value.. Modes: Text to Image, Image Edit
- `enable_safety_checker` (boolean): If set to true, the safety checker will be enabled.. Modes: Text to Image, Image Edit. Only available via fal
- `height` (integer): Height of output image. Modes: Text to Image, Image Edit. Only available via replicate
- `image_2` (string): Optional second input image for multi-image operations. Modes: Text to Image, Image Edit. Only available via replicate
- `image_3` (string): Optional third input image for multi-image operations. Modes: Text to Image, Image Edit. Only available via replicate
- `image_guidance_scale` (number): The Image Guidance scale controls how closely the model follows the input images. For image editing: 1.3-2.0, for in-context generation: 2.0-3.0. Modes: Text to Image, Image Edit
- `max_input_image_side_length` (integer): Maximum input image side length. Modes: Text to Image, Image Edit. Only available via replicate
- `max_pixels` (integer): Maximum number of pixels in output. Modes: Text to Image, Image Edit. Only available via replicate
- `num_inference_steps` (integer): The number of inference steps to perform.. Modes: Text to Image, Image Edit
- `scheduler` (string): The scheduler to use for the diffusion process.. Values: dpmsolver, euler. Modes: Text to Image, Image Edit
- `sync_mode` (boolean): If `True`, the media will be returned as a data URI and the output data won't be available in the request history.. Modes: Text to Image, Image Edit. Only available via fal
- `text_guidance_scale` (number): The Text Guidance scale controls how closely the model follows the text prompt. Higher values make the model stick more closely to the prompt.. Modes: Text to Image, Image Edit
- `width` (integer): Width of output image. Modes: Text to Image, Image Edit. Only available via replicate


## Model Identifiers

- Primary Slug: omnigen-v2
- Aliases: omnigen2

## Dates

- Released: June 2025

## Tags

image-generation, text-to-image, image-editing

## Available Providers

### fal.ai

- Config Key: fal/omnigen-v2
- Provider Model ID: fal-ai/omnigen-v2
- Pricing: $0.150/megapixel
  - Source: https://fal.ai/models/fal-ai/omnigen-v2

### Replicate

- Config Key: replicate/omnigen-v2
- Provider Model ID: lucataco/omnigen2
- Pricing: $0.088/image
  - Note: Community model, pricing is hardware-based (Nvidia L40S)
  - Note: ~$0.088 per run at typical completion time of ~91 seconds
  - Source: https://replicate.com/lucataco/omnigen2


## Arena Benchmarks

### Geometric Composition

- Elo: 1124
- Record: 1W / 5L / 0T (6 battles)
- Rank: #49 of 70

### Candid Street Photography

- Elo: 1117
- Record: 1W / 3L / 1T (5 battles)
- Rank: #42 of 68

### Magic Burger Explosion: Fiery Photorealism Challenge

- Elo: 1108
- Record: 1W / 4L / 0T (5 battles)
- Rank: #46 of 69

### Isometric Miniature Diorama Scenes

- Elo: 1079
- Record: 1W / 2L / 1T (4 battles)
- Rank: #59 of 72

### Vintage Cafe Logo

- Elo: 1057
- Record: 1W / 6L / 1T (8 battles)
- Rank: #60 of 70

### Fantasy Warrior

- Elo: 1054
- Record: 1W / 4L / 1T (6 battles)
- Rank: #60 of 71

### The Halloween Invitation

- Elo: 1035
- Record: 1W / 4L / 0T (5 battles)
- Rank: #64 of 70

### Adorable Baby Animals in Sunny Meadow

- Elo: 1010
- Record: 0W / 8L / 0T (8 battles)
- Rank: #68 of 72


## Use Cases & Category Performance

### Creativity (Text-to-Image)

- Rank: #33 of 62
- Elo: 1211
- Record: 5W / 19L / 4T (28 battles)
- Win Rate: 17.9%

### 3D Imaging & Modeling (Text-to-Image)

- Rank: #42 of 61
- Elo: 1169
- Record: 2W / 7L / 1T (10 battles)
- Win Rate: 20.0%

### Text Rendering (Text-to-Image)

- Rank: #46 of 61
- Elo: 1194
- Record: 5W / 22L / 3T (30 battles)
- Win Rate: 16.7%

### Portraits (Text-to-Image)

- Rank: #46 of 50
- Elo: 1034
- Record: 1W / 4L / 1T (6 battles)
- Win Rate: 16.7%

### Cartoon, Anime & Fantasy (Text-to-Image)

- Rank: #46 of 50
- Elo: 1034
- Record: 1W / 4L / 1T (6 battles)
- Win Rate: 16.7%

### Aesthetics (Text-to-Image)

- Rank: #58 of 62
- Elo: 1170
- Record: 10W / 41L / 4T (55 battles)
- Win Rate: 18.2%

### Product, Branding & Commercial (Text-to-Image)

- Rank: #58 of 61
- Elo: 1148
- Record: 4W / 20L / 2T (26 battles)
- Win Rate: 15.4%

### Art (Text-to-Image)

- Rank: #55 of 57
- Elo: 1095
- Record: 2W / 5L / 0T (7 battles)
- Win Rate: 28.6%

### Photorealism (Text-to-Image)

- Rank: #60 of 62
- Elo: 1108
- Record: 6W / 29L / 2T (37 battles)
- Win Rate: 16.2%

### Prompt Adherence (Text-to-Image)

- Rank: #61 of 62
- Elo: 1150
- Record: 6W / 28L / 2T (36 battles)
- Win Rate: 16.7%


## Image Gallery

12 images available for this model. Browse all at https://lumenfall.ai/models/vectorspacelab/omnigen-v2/gallery

### Curated Examples

- [A cinematic, wide-angle shot of a high-end, contemporary boutique storefront at dusk. The shop’s ...](https://assets.lumenfall.ai/81SO0oOdtOzJC9i98GooGd0oR_hcs9bp35zQnMH_M4s/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/5orf0jk1gagbxp0352m473tls0k9@jpeg)
- [A hyper-realistic close-up portrait of an elderly artisan woodworker in a sun-drenched workshop. ...](https://assets.lumenfall.ai/Bi9h_I7NiOK_4_sYCccV38Ecx0z4PtFKP_1Qlzqd9dg/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/c9py0le3lv2btp6bwwmiwmg7wipz@jpeg)
- [Cinematic close-up portrait of an elderly artisan watchmaker in a sun-drenched attic workshop, me...](https://assets.lumenfall.ai/H7bFOy1_z6bLlgS5TQibBs7dETU-azsb3rFVNaXQYGU/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/ehy8arusuifqhq4lim1a5mmm4o64@jpeg)
- [A sunlit, cozy bookstore window with a gold leaf sign on the glass that reads "READ MORE BOOKS" i...](https://assets.lumenfall.ai/uPc48cXmbXMkC7zQLBEICqrTiGOhP-ufAhH_tc_un3c/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/mp0kdecfes3nnkqq4mamt6kyxqic@jpeg)

### Arena Competition Results

- [Geometric Composition](https://assets.lumenfall.ai/srk1naiWISnz6dbX1boUlzDeZ2V9tff3l8_Ck0TO7dc/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/o98uknb0n4mo9uhpyk0twupce0bk@jpeg): #50 of 70 (Elo 1124)
- [Candid Street Photography](https://assets.lumenfall.ai/DasgjMO2wGdesqtiXMKAXr1ycxVf3dN37aGQkFHDbOI/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/fcrr446hll065prw6ir0drr2xfc1@jpeg): #42 of 68 (Elo 1117)
- [Magic Burger Explosion: Fiery Photorealism Challenge](https://assets.lumenfall.ai/_tsH8UtmT4NpYyv9iHFy3TbI01S7Jl_vODcpRex0HPY/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/4l8ajqvb4owbu5necz9bk7cqvd2r@jpeg): #46 of 69 (Elo 1107)
- [Isometric Miniature Diorama Scenes](https://assets.lumenfall.ai/3Mx7Tmqvc0XYaBy8BXXpxTaOa-7TjaVa4JzjNAKkHYQ/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/s1ylpsur6q6pf2xpby87gxk9vsrc@jpeg): #59 of 72 (Elo 1079)
- [Vintage Cafe Logo](https://assets.lumenfall.ai/fBF7FRcLWQDbeAPjRZq89WnccQ_IenYG1YO8AFbdi5c/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/i9n76bq4tdkf23k7oc22hut60iqo@jpeg): #61 of 70 (Elo 1057)
- [Fantasy Warrior](https://assets.lumenfall.ai/JkknYvgsATfMmAs66I50oghLxCjfQLmviUUObTO1oI0/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/tnkirf73q4k4ydq36vg68g6fe0zx@jpeg): #61 of 71 (Elo 1054)
- [The Halloween Invitation](https://assets.lumenfall.ai/B5bfpWaDlR-F2k25Mw9QUKVo2C1bH62IYgVZvtD2Qhk/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/6d7xtg01xc3rf7g63ujpi9tnr41v@jpeg): #64 of 70 (Elo 1035)
- [Adorable Baby Animals in Sunny Meadow](https://assets.lumenfall.ai/o3U6o33PKgzU38Yn5722NJTfLXxKFAXgGezZtxD5lGk/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/gyburvrurbtdisc3wnyl1svapnr9@jpeg): #68 of 72 (Elo 1010)

## Example Prompt

The following prompt was used to generate an example image in our playground:

A sunlit, cozy bookstore window with a gold leaf sign on the glass that reads "READ MORE BOOKS" in elegant serif typography. Inside, a woman browses the shelves. On the sidewalk bench outside, a capybara sits quietly watching the people pass by.

## Code Examples

### Text to Image (/v1/images/generations)

#### cURL

curl -X POST \
  https://api.lumenfall.ai/openai/v1/images/generations \
  -H "Authorization: Bearer $LUMENFALL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "omnigen-v2",
    "prompt": "",
    "size": "1024x1024"
  }'

# Response:
# { "created": 1234567890, "data": [{ "url": "https://...", "revised_prompt": "..." }] }

#### JavaScript

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://api.lumenfall.ai/openai/v1'
});

const response = await client.images.generate({
  model: 'omnigen-v2',
  prompt: '',
  size: '1024x1024'
});

// { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }
console.log(response.data[0].url);

#### Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.lumenfall.ai/openai/v1"
)

response = client.images.generate(
    model="omnigen-v2",
    prompt="",
    size="1024x1024"
)

# { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }
print(response.data[0].url)

### Image Edit (/v1/images/edits)

#### cURL

curl -X POST \
  https://api.lumenfall.ai/openai/v1/images/edits \
  -H "Authorization: Bearer $LUMENFALL_API_KEY" \
  -F "model=omnigen-v2" \
  -F "image=@source.png" \
  -F "prompt=Add a starry night sky to this image" \
  -F "size=1024x1024"

# Response:
# { "created": 1234567890, "data": [{ "url": "https://...", "revised_prompt": "..." }] }

#### JavaScript

import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://api.lumenfall.ai/openai/v1'
});

const response = await client.images.edit({
  model: 'omnigen-v2',
  image: fs.createReadStream('source.png'),
  prompt: 'Add a starry night sky to this image',
  size: '1024x1024'
});

// { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }
console.log(response.data[0].url);

#### Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.lumenfall.ai/openai/v1"
)

response = client.images.edit(
    model="omnigen-v2",
    image=open("source.png", "rb"),
    prompt="Add a starry night sky to this image",
    size="1024x1024"
)

# { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }
print(response.data[0].url)


## Frequently Asked Questions

### How much does OmniGen v2 cost?

OmniGen v2 starts at $0.088 per image through Lumenfall. Pricing varies by provider. Lumenfall does not add any markup to provider pricing.

### How do I use OmniGen v2 via API?

You can use OmniGen v2 through Lumenfall's OpenAI-compatible API. Send requests to the unified endpoint with model ID "omnigen-v2". Code examples are available in Python, JavaScript, and cURL.

### Which providers offer OmniGen v2?

OmniGen v2 is available through fal.ai and Replicate on Lumenfall. Lumenfall automatically routes requests to the best available provider.

## Links

- Model Page: https://lumenfall.ai/models/vectorspacelab/omnigen-v2
- About: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/about
- Providers, Pricing & Performance: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/providers
- API Reference: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/api
- Benchmarks: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/benchmarks
- Use Cases: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/use-cases
- Gallery: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/gallery
- Playground: https://lumenfall.ai/models/vectorspacelab/omnigen-v2/playground
- API Documentation: https://docs.lumenfall.ai