ACS Email now Generally Available in Azure Government
August 5, 2025
Blocking risky users with Passwordless Authentication methods
August 5, 2025
We’re excited to announce Azure AI Foundry Models now hosts FLUX.1 Kontext [pro] and FLUX1.1 [pro] as direct from Azure, giving developers a first-party, enterprise-ready path to Black Forest Labs’ (BFL) state-of-the-art image models. You get secure endpoints with pay-as-you-go model, Azure billing and Content Safety integration—no GPU wrangling required.
Meet the Models
|
Model |
Core task |
What’s new |
Speed |
Resolution / IO |
|
FLUX.1 Kontext [pro] |
in-context image generation and editing (text + image prompt) |
single model unifies local edits, full scene re-gen, style transfer, character consistency, iterative editing |
up to 8× faster than other SOTA editors |
1024 x 1024 default; iterative multi-turn editing |
|
FLUX1.1 [pro] |
text-to-image |
Ultra mode: 4 MP images, Raw mode for natural “camera” look |
6× faster than Flux 1-pro; 10 s for a 4 MP frame |
up to 4 MP, strong prompt adherence |
Under the hood: Both models sit on a rectified flow transformer backbone—BFL’s answer to diffusion and latent consistency models—yielding better sample diversity and lower inference latency.
Image Capabilities & Enterprise Use-case Patterns
Exploring the power of FLUX.1 Kontext [pro], we put its in-context image generation and editing capabilities to the test in Azure AI Foundry, transforming simple prompts into stunning, detailed visuals that showcase just how far generative AI has come.
Prompt 1: “Two children sailing a paper boat down a winding river, surrounded by lush jungles and curious animals”
Prompt 2: “Abstract digital painting of a futuristic city at sunset, with glowing neon lights and flying vehicles, in cyberpunk style”
Prompt 3: “Surreal landscape made of floating islands, waterfalls spilling into the sky, and glowing crystal trees”
With these Black Forest Labs models now available on Azure AI Foundry, enterprises are enabled to accelerate creative pipelines, generate e-commerce variants, automate marketing workflows and simulate digital twins at scale.
|
Scenario |
Pattern to Try |
|
Creative Pipeline Acceleration |
Use FLUX 1.1 [pro] for storyboard ideation → pass frames into Kontext [pro] for surgical tweaks without PSD layers. |
|
E-commerce Variant Generation |
Inject product hero shot + prompt to FLUX.1 Kontext [pro] to auto-paint seasonal backdrops while preserving SKU angles. |
|
Marketing Automation |
Pair Azure OpenAI GPT-4o for copy + FLUX images via Logic Apps; send variants to A/B email testing. |
|
Digital Twin Simulation |
Use iterative editing to visualize wear/tear on equipment over time in maintenance portals. |
Benchmarks & Economics
- Latency: FLUX.1 Kontext [pro] averages 0.9 s per 1024 x 1024 edit—eight times faster than leading diffusion-based editors on identical A100s.
- Quality: On KontextBench, FLUX.1 Kontext [pro] ranks #1 on text-guided editing and character-consistency, while FLUX 1.1 [pro] tops aesthetics and prompt-following in T2I tests.
Pricing
|
Model Name |
Meter Type |
Price |
|
FLUX 1.1 [pro] Global |
1K Images |
$40 |
|
FLUX.1 Kontext [pro] Global |
1K Images |
$40 |
Tips for Production Readiness
- Seed for determinism: Both models accept seed for repeatable outputs—store alongside prompt history.
- Step budget: Ultra-mode images look best with 40-50 inference steps; FLUX.1 Kontext [pro] edits converge in < 30.
- Guard-rail chaining: Pipe outputs through Azure AI Content Safety and your own watermark classifier.
- Caching: For high-traffic apps, cache intermediate latent representations (Kontext) to speed multi-turn edits.
Why Azure AI Foundry?
Direct from Azure models give you the fastest time-to-value on cutting-edge foundation models, while Azure AI Foundry supplies the right tools, evaluation, deployment, safety, and lifecycle plumbing needed by real-world enterprises.
|
What You Get |
Why It Matters |
|
Unified access |
Switch or chain models without rewriting code or juggling separate keys/resources. |
|
Enterprise-ready SLAs & security |
Meets compliance officers where they live—no third-party contracts, guaranteed uptime. |
|
Scalable deployments |
Start small in dev, flip to prod traffic without re-deploying. |
|
Deep toolchain hook-ups |
Shorter path from hack-day demo to governed production workload. |
Build Trustworthy AI Solutions
Black Forest Labs models on Azure AI Foundry are delivered under the Microsoft Product Terms, giving you enterprise-grade security and compliance out of the box. Each FLUX endpoint offers secure Content Safety controls and guardrails. Runtime protections include built-in content-safety filters, role-based access control, virtual-network isolation, and automatic Azure Monitor logging. Governance signals stream directly into Azure Policy, Purview, and Microsoft Sentinel, giving security and compliance teams real-time visibility. Together, Microsoft’s capabilities let you create with more confidence, knowing that privacy, security, and safety are woven into every Black Forest Labs deployment from day one.
How to Deploy BFL Models in Azure AI Foundry?
- If you don’t have an Azure subscription, you can sign up for an Azure account here.
- Search for the model name in the model catalog in Azure AI Foundry.
- FLUX.1-Kontext-pro
- FLUX-1.1-pro
- Open the model card in the model catalog.
- Click on deploy to obtain the inference API and key and also to access the playground.
- You should land on the deployment page that shows you the API and key in less than a minute. You can try out your prompts in the playground.
- You can use the API and key with various clients.
The FLUX family has already re-defined speed/quality trade-offs in open image generation. Landing FLUX.1 Kontext [pro] and FLUX 1.1 [pro] inside Azure AI Foundry brings those capabilities—with Azure’s scalability, governance, and integrated tooling—to every developer building imaging workflows. Happy generating!
Learn More
▶️ RSVP for the next Model Monday LIVE on YouTube or On-Demand
👩💻 Explore Azure AI Foundry Models
👋 Continue the conversation on Discord