# FLUX.2 [pro]

> Black Forest Labs' state-of-the-art image generation model with maximum quality and speed, supporting text-to-image and multi-reference image editing with up to 4MP output

## Quick Reference

- Model ID: flux.2-pro
- Creator: Black Forest Labs
- Status: active
- Family: flux.2
- Base URL: https://api.lumenfall.ai/openai/v1

## Specifications

- Max Resolution: 2048x2048
- Max Input Images: 9
- 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/flux.2-pro?schema=true`.

### Core Parameters

- `prompt` (string) — REQUIRED: Edit instruction for the image. Modes: Image Edit, Text to Image
- `seed` (integer): Random seed for reproducibility. Modes: Image Edit, Text to Image

### Size & Layout

- `size` (string): Image dimensions as WxH pixels (e.g. "1024x1024") or aspect ratio (e.g. "16:9"). Values: auto, 1365x768, 768x1365, 1254x836, 836x1254, 887x1182, 1024x1024, 1145x916, 916x1145, 1183x887. Modes: Image Edit, Text to Image
- `aspect_ratio` (string): Aspect ratio of the output image (e.g. "16:9", "1:1"). Values: auto, 9:16, 2:3, 3:4, 4:5, 1:1, 5:4, 4:3, 3:2, 16:9. Modes: Image Edit, Text to Image
- `resolution` (string): Output resolution tier (e.g. "1K", "4K"). Values: auto, 1K. Modes: Image Edit, Text to Image

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

### Additional Parameters

- `enable_safety_checker` (boolean): Whether to enable the safety checker.. Modes: Image Edit, Text to Image. Only available via fal
- `height` (integer): Height of the generated image. Only used when aspect_ratio=custom. Must be a multiple of 16 (if it's not, it will be rounded to nearest multiple of 16).. Modes: Text to Image, Image Edit. Only available via replicate
- `output_quality` (integer): Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs. Modes: Text to Image, Image Edit. Only available via replicate
- `safety_tolerance` (string): The safety tolerance level for the generated image. 1 being the most strict and 5 being the most permissive.. Values: 1, 2, 3, 4, 5. Modes: Image Edit, Text to Image
- `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: Image Edit, Text to Image. Only available via fal
- `width` (integer): Width of the generated image. Only used when aspect_ratio=custom. Must be a multiple of 16 (if it's not, it will be rounded to nearest multiple of 16).. Modes: Text to Image, Image Edit. Only available via replicate


## Model Identifiers

- Primary Slug: flux.2-pro
- Aliases: flux-2-pro


## Tags

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

## Available Providers

### fal.ai

- Config Key: fal/flux.2-pro-edit
- Provider Model ID: fal-ai/flux-2-pro/edit
- Pricing: $0.015/megapixel, 
  - Note: Resolution is rounded up to the next megapixel, separately for each reference image and the generated image
  - Note: 1 megapixel = 1024x1024 pixels
  - Note: Each reference image is counted separately (minimum 1 MP each)
  - Note: Images exceeding 4 megapixels are resized to 4 megapixels
  - Source: https://fal.ai/models/fal-ai/flux-2-pro

### fal.ai

- Config Key: fal/flux.2-pro
- Provider Model ID: fal-ai/flux-2-pro
- Pricing: 
  - Note: Resolution is rounded up to the next megapixel
  - Note: 1 megapixel = 1024x1024 pixels
  - Source: https://fal.ai/models/fal-ai/flux-2-pro

### Replicate

- Config Key: replicate/flux.2-pro
- Provider Model ID: black-forest-labs/flux-2-pro
- Pricing: $0.015/request, $0.015/megapixel, $0.015/megapixel
  - Note: Resolution is rounded up to the next megapixel, separately for each reference image and the generated image
  - Note: 1 megapixel = 1024x1024 pixels
  - Note: Each reference image is counted separately (minimum 1 MP each)
  - Note: Images exceeding 4 megapixels are resized to 4 megapixels
  - Source: https://replicate.com/black-forest-labs/flux-2-pro


## Performance Metrics

Provider performance over the last 30 days.

### replicate

- Median Generation Time (p50): 14207ms
- 95th Percentile Generation Time (p95): 28969ms
- Average Generation Time: 18177ms
- Success Rate: 100.0%
- Total Requests: 22
- Time to First Byte (p50): 13710ms
- Time to First Byte (p95): 28363ms

### fal

- Median Generation Time (p50): 16771ms
- 95th Percentile Generation Time (p95): 23819ms
- Average Generation Time: 18307ms
- Success Rate: 100.0%
- Total Requests: 20
- Time to First Byte (p50): 16412ms
- Time to First Byte (p95): 21311ms


## Arena Benchmarks

### Man and Car in California

- Elo: 1259
- Record: 38W / 50L / 5T (93 battles)
- Rank: #7 of 13

### Night Sky Transformation

- Elo: 1241
- Record: 13W / 16L / 5T (34 battles)
- Rank: #4 of 16

### Studio Ghibli Anime Style

- Elo: 1231
- Record: 17W / 10L / 0T (27 battles)
- Rank: #5 of 14

### Vintage Cafe Logo

- Elo: 1230
- Record: 16W / 8L / 0T (24 battles)
- Rank: #3 of 21

### Over-the-top cartoon caricature

- Elo: 1208
- Record: 16W / 11L / 2T (29 battles)
- Rank: #6 of 13

### Fantasy Warrior

- Elo: 1199
- Record: 11W / 4L / 0T (15 battles)
- Rank: #11 of 21

### Heroic Super Hero Portrait

- Elo: 1192
- Record: 17W / 4L / 5T (26 battles)
- Rank: #10 of 21

### Modern Clean Menu

- Elo: 1181
- Record: 9W / 12L / 1T (22 battles)
- Rank: #12 of 19

### Adorable Baby Animals in Sunny Meadow

- Elo: 1179
- Record: 8W / 6L / 0T (14 battles)
- Rank: #11 of 25

### Golden Hour Stroll

- Elo: 1173
- Record: 11W / 22L / 0T (33 battles)
- Rank: #10 of 13

### Neutral Expression to Genuine Smile

- Elo: 1170
- Record: 13W / 22L / 2T (37 battles)
- Rank: #13 of 14

### Bald man challenge

- Elo: 1165
- Record: 11W / 20L / 1T (32 battles)
- Rank: #12 of 15

### Candid Street Photography

- Elo: 1157
- Record: 7W / 6L / 0T (13 battles)
- Rank: #17 of 24

### Geometric Composition

- Elo: 1155
- Record: 9W / 9L / 3T (21 battles)
- Rank: #21 of 22

### Apollo 11: Journey to Tranquility

- Elo: 1153
- Record: 9W / 13L / 1T (23 battles)
- Rank: #13 of 19

### Fantasy Warrior

- Elo: 1147
- Record: 5W / 5L / 0T (10 battles)
- Rank: #2 of 14


## Use Cases & Category Performance

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

- Rank: #2 of 21
- Elo: 1244
- Record: 15W / 7L / 0T (22 battles)
- Win Rate: 68.2%

### Anime (Image Editing)

- Rank: #5 of 14
- Elo: 1216
- Record: 17W / 10L / 0T (27 battles)
- Win Rate: 63.0%

### Portrait (Text-to-Image)

- Rank: #9 of 21
- Elo: 1207
- Record: 11W / 4L / 0T (15 battles)
- Win Rate: 73.3%

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

- Rank: #10 of 23
- Elo: 1230
- Record: 33W / 32L / 2T (67 battles)
- Win Rate: 49.3%

### Photorealism (Text-to-Image)

- Rank: #12 of 24
- Elo: 1195
- Record: 7W / 6L / 0T (13 battles)
- Win Rate: 53.8%

### Photorealism (Image Editing)

- Rank: #9 of 16
- Elo: 1222
- Record: 86W / 130L / 13T (229 battles)
- Win Rate: 37.6%

### Portrait (Image Editing)

- Rank: #11 of 15
- Elo: 1193
- Record: 24W / 42L / 3T (69 battles)
- Win Rate: 34.8%


## Image Gallery

20 images available for this model. Browse all at https://lumenfall.ai/models/black-forest-labs/flux.2-pro/gallery

### Curated Examples

- [A wide, cinematic shot of a sophisticated, high-end camera lens boutique located on a quiet Europ...](https://assets.lumenfall.ai/l_suHtKjWQlhRqzSWTpxXavohfFDyVXo-t1_xX0PVsU/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/28j4xuwljif9wzr1znt924ayz2jm@jpeg)
- [A medium shot of an elderly artisan in a sun-drenched Mediterranean workshop, meticulously carvin...](https://assets.lumenfall.ai/4sGkyeWS1cerhMKjYtiaDgNj8iZjYMgAtw4otfwq_JM/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/dttukfw2q8i4q4chxmi9zdzyl6x0@jpeg)
- [A medium shot of an elderly artisan in a sun-drenched Tuscan workshop, meticulously hand-carving ...](https://assets.lumenfall.ai/DZLcBLL4SZUrKxUyJRO2GO9EnFo_z6DQsa27LtXsKFQ/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/4uyh7h4vfp3rba8wqgyow0hz0d1n@jpeg)
- [A hyper-realistic close-up of a rustic wooden sign outside a cozy bookstore that reads "THE LATE ...](https://assets.lumenfall.ai/pj3HTLidP1tdSe1gnAlQamQi6duq3cqaElnNCc9m3Xg/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/14t2cx7rduu0hfw7unat7m0a59tt@jpeg)

### Arena Competition Results

- [Man and Car in California](https://assets.lumenfall.ai/YVX7s4_EoHvTdkcc8PxJkFeWpB8DKlFvPJa1J1FvPSs/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/jhn3buil59ve9qun8pix3fkcrq3g@jpeg): #7 of 13 (Elo 1259)
- [Night Sky Transformation](https://assets.lumenfall.ai/CLLRoyOAVQOBbSqrpj9Vqp6bEei8ySSO_USDtQTq2CU/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/ao6qtidb9bbkmn4nycebzr50njvu@jpeg): #4 of 16 (Elo 1241)
- [Studio Ghibli Anime Style](https://assets.lumenfall.ai/LmbQK45YLZuzCOSLuUMLgzcybD7kFO8UoARUdUE9PzQ/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/5trj5bwaaw02zxfufop3f72ct0g1@jpeg): #5 of 14 (Elo 1231)
- [Vintage Cafe Logo](https://assets.lumenfall.ai/srVniR94Efu4LZlWn1BrXqDdlzGfU9V-qdRd4JgQzo8/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/g48kusernizcfih5iidknlf05i4z@jpeg): #3 of 21 (Elo 1230)
- [Over-the-top cartoon caricature](https://assets.lumenfall.ai/QhyrdG5R_Nw4HjdYUwu_K9rZq4h_S9mYC-gH8Gv19S8/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/2ac5h1lpe41u7idb8cd2dy7iar8h@jpeg): #6 of 13 (Elo 1208)
- [Fantasy Warrior](https://assets.lumenfall.ai/sPTh8csdEADPgjJ0RS0krz5bB1xp1RR9q0dhsXHYXqs/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/g28typ55ul03ess25bh7biobg1bh@jpeg): #11 of 21 (Elo 1199)
- [Heroic Super Hero Portrait](https://assets.lumenfall.ai/kDu3zpxDTxSXOpYsHPG6mzC2LBYdt9fJFWh_Ptt6118/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/pyt68w2b7h0te6hc7ddjk5b34cca@jpeg): #10 of 21 (Elo 1192)
- [Modern Clean Menu](https://assets.lumenfall.ai/S11YSqWWpGLRlQZNq33Yqlosk6gWE9noYEeJi0RdGQk/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/tlvp5fr5u557mpopku17znzu0hb4@jpeg): #12 of 19 (Elo 1181)
- [Adorable Baby Animals in Sunny Meadow](https://assets.lumenfall.ai/n3mogu9L0voG8dzRqdArd0pnVeR1O5A-iYfo9BC5c-A/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/lu64vcq9htctichtarv5jemycvkd@jpeg): #11 of 25 (Elo 1179)
- [Golden Hour Stroll](https://assets.lumenfall.ai/bQOcu0luG84sDwlmzpL7MYwDXPHWeXnpOrx2rhznJe8/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/gy0duj27dtcmerfa4hgqq9bh7qdz@jpeg): #10 of 13 (Elo 1173)
- [Neutral Expression to Genuine Smile](https://assets.lumenfall.ai/ucZSbJhwMNi4hlStdINRiwaQvQ_4TZvNvahbXROcs-A/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/qqev4ly3o0d7l4cxbvl7x2j60sus@jpeg): #13 of 14 (Elo 1170)
- [Bald man challenge](https://assets.lumenfall.ai/wDVv61FxxfyHtwkfKPH5NqTqg3CFIKWFZt5X9hku05s/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/7vypr2wmfrb33vggpy87f5tpg3w7@jpeg): #12 of 15 (Elo 1165)
- [Candid Street Photography](https://assets.lumenfall.ai/cEvxIi1kjv_LOpkeq0e--TU91_iqsCWTPJpcyUQZrKQ/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/xe84gsk4wuxi42lqh8ym0rak8ql1@jpeg): #17 of 24 (Elo 1157)
- [Geometric Composition](https://assets.lumenfall.ai/0As0J51QfVuvjTGEv0ezY6-3fUpFnNdQyW4yoyXErx4/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/czexic1nlmh6x2m8xecd7pnlmnyf@jpeg): #21 of 22 (Elo 1155)
- [Apollo 11: Journey to Tranquility](https://assets.lumenfall.ai/2SCM_Z3EzEptm6yLgxAB6IrJ0xJ1QV_-otvrWIMLjv4/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/8wh5tmgjl0p2mpdrq2a0guywqh2x@jpeg): #13 of 19 (Elo 1153)
- [Fantasy Warrior](https://assets.lumenfall.ai/3JYF_tnVo8GxLrQUfN0QtuF5KwZHoHW2HavcZOk5L80/rs:fit:1500:1500/plain/gs://lumenfall-prod-assets/zefz837kpcam0w18t262in7g78tr@jpeg): #2 of 14 (Elo 1147)

## Example Prompt

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

A hyper-realistic close-up of a rustic wooden sign outside a cozy bookstore that reads "THE LATE CHAPTER" in elegant gold leaf typography. In the soft-focus background, a capybara is calmly sitting on a park bench next to a reading lamp.

## 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": "flux.2-pro",
    "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: 'flux.2-pro',
  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="flux.2-pro",
    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=flux.2-pro" \
  -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: 'flux.2-pro',
  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="flux.2-pro",
    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)


## About

## Overview
FLUX.2 [pro] is the flagship high-performance image generation model developed by Black Forest Labs. It is designed to produce high-fidelity visuals up to 4 megapixels while maintaining a balance between aesthetic quality and inference speed. A distinguishing feature of this iteration is its native support for multi-reference image editing, allowing users to guide generation using existing visual assets alongside text prompts.

## Strengths
*   **High-Resolution Output:** Capable of generating detailed images up to 4MP, providing significantly more pixel-level data than standard 1MP models for large-format applications.
*   **Multi-Reference Consistency:** Excels at "image-to-image" and reference-based workflows, using multiple source images to maintain consistency in style, character, or composition during the editing process.
*   **Complex Prompt Adherence:** Demonstrates high precision in following intricate text instructions, particularly when handling spatial relationships between objects and specific lighting conditions.
*   **Text Rendering Accuracy:** Inherits the family’s capability for rendering legible, correctly spelled text within generated images, even in complex fonts or curved layouts.

## Limitations
*   **Computational Cost:** As the "pro" tier model, it carries a higher price point per generation ($0.015) compared to the "schnell" or "dev" variants, making it less ideal for high-volume rapid prototyping.
*   **Latency Tradeoff:** While optimized for speed relative to its output size, the sheer volume of pixels (4MP) results in longer generation times than lower-resolution, distilled models.
*   **Hardware Requirements:** Due to its scale and state-of-the-art weights, it is generally restricted to managed API environments rather than consumer-grade local hardware.

## Technical Background
FLUX.2 [pro] is built on a scaled diffusion transformer architecture, a refinement of the original Flow-based models developed by the core team behind Stable Diffusion. The training approach focuses on maximizing the signal-to-noise ratio at high resolutions, utilizing a "pro" weight set that has been fine-tuned for photorealism and professional-grade color accuracy. It utilizes advanced latent space compression to handle the 4MP output without proportional increases in VRAM usage.

## Best For
This model is best suited for professional design workflows, advertising photography, and high-end digital art where resolution and prompt fidelity are non-negotiable. It is particularly effective for brand-consistent content creation where multiple reference images must define the output's look and feel. FLUX.2 [pro] is available for testing and deployment through Lumenfall's unified API and interactive playground, allowing developers to integrate its high-resolution capabilities into their applications with minimal overhead.

## Frequently Asked Questions

### How much does FLUX.2 [pro] cost?

FLUX.2 [pro] starts at $0.015 per image through Lumenfall. Pricing varies by provider. Lumenfall does not add any markup to provider pricing.

### How do I use FLUX.2 [pro] via API?

You can use FLUX.2 [pro] through Lumenfall's OpenAI-compatible API. Send requests to the unified endpoint with model ID "flux.2-pro". Code examples are available in Python, JavaScript, and cURL.

### Which providers offer FLUX.2 [pro]?

FLUX.2 [pro] is available through fal.ai and Replicate on Lumenfall. Lumenfall automatically routes requests to the best available provider.

### What is the maximum resolution for FLUX.2 [pro]?

FLUX.2 [pro] supports images up to 2048x2048 resolution.

## Links

- Model Page: https://lumenfall.ai/models/black-forest-labs/flux.2-pro
- About: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/about
- Providers, Pricing & Performance: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/providers
- API Reference: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/api
- Benchmarks: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/benchmarks
- Use Cases: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/use-cases
- Gallery: https://lumenfall.ai/models/black-forest-labs/flux.2-pro/gallery
- Playground: https://lumenfall.ai/playground?model=flux.2-pro
- API Documentation: https://docs.lumenfall.ai