AI on Demandpowered by Black Forest Labs

FLUX.1-schnell is a state-of-the-art text-to-image model designed for fast image generation. It is openly available under the Apache 2.0 license, making it suitable for a wide range of commercial and research applications.

FLUX.1-fast is a powerful open-weight model for generating high-quality images from natural language descriptions. Despite its comparatively small size of 12 billion parameters, the model delivers first-class image quality whilst requiring only 1 to 4 inference steps. As a result, it achieves a level of speed and quality comparable to leading proprietary text-to-image models. The open model architecture also allows for flexible use within users’ own applications and infrastructures.

FLUX.1-schnell is ideal for developers, businesses and creative professionals who wish to integrate image generation efficiently into their applications. Typical use cases include the creation of marketing materials, product visualisations, illustrations, concept art and AI-powered creative tools. Thanks to its low latency, the model is particularly well-suited to real-time applications, interactive user interfaces and productive workflows.

The combination of high image quality and extremely fast generation makes FLUX.1-schnell a powerful solution for modern AI applications. As an open-weight model, it can be seamlessly integrated into existing systems. The low computational overhead, resulting from just a few inference steps, reduces operating costs and enables the efficient use of existing GPU resources.

Your Consultant

Scope of services

On-demand AI image generation

Quick access to FLUX.1 for the rapid generation of high-quality images from text descriptions. The model delivers convincing results after just 1 to 4 inference steps.

Efficient GPU performance

Optimised GPU resources for fast response times and high throughput. The 12-billion-parameter architecture combines high image quality with efficient computational overhead.

Managed service

Deployment, monitoring and maintenance on Swiss infrastructure, including personalised support. stepping stone takes care of operations so that you can focus on your applications.

Areas of application

Marketing & Content Creation

FLUX.1 quickly generates high-quality images for campaigns, social media, websites and presentations directly from text descriptions.

Companies use the model to create bespoke graphics in a short space of time, accelerate creative processes and scale content efficiently.

Product Design & AI Applications

FLUX.1 quickly generates concept graphics, product visualisations and design variants within a few seconds and integrates seamlessly into existing applications.

From prototyping and personalised image generation to automated creative workflows, companies benefit from low latency, high image quality and flexible deployment on Swiss infrastructure.

Benchmark

The benchmarks were measured with a python script.
For a fixed resolution the time per image is measured, which is needed for generating the image.
Because the required time depends on resolution and number of steps different configurations can be tested.
Step-by-step instructions and the required Python script can be downloaded from GitHub.

Call

# Set your personal key:
STONEY_KEY=sk-...

# Make key visible for bench script:
export OPENAI_API_KEY=$STONEY_KEY

# Start the benchmark
python flux_bench.py \
  --base-url llm.stoney-cloud.com \
  --model black-forest-labs/FLUX.1-schnell \
  --num-images 20 \
  --size 1024x1024 \
  --steps 4 \

Result

====================================================
 FLUX.1 [schnell] benchmark results
====================================================
 config:          1024x1024, 4 steps
 megapixels:      1.049 MP
 images ok:       20
 images failed:   0
 wall-clock:      31.88s
 mean latency:     1.59s / image
 p99:              1.75s
 min / max:        1.48s / 1.75s
 throughput:      37.6 images / minute
                   0.63 images / second
====================================================

Legend

  • config: The configuration of the image used in the request (size and steps).
  • megapixels: The resolution of the images in megapixels.
  • images ok: The number of successfully generated images.
  • images failed: The number of unsuccessfully generated images.
  • wall-clock: The duration of the benchmark.
  • mean latency: The average duration of an image generation.
  • p99: The time needed for an image generation in the worst case.
  • min / max: The minimal and maximal duration of an image generation.
  • throughput: The number of images generated per second and minute given the configuration.

Price

ModelPrompt lengthOutput/Megapixel
FLUX.1-schnell5120.0030
All prices are in CHF/MTok, excluding VAT.