8 posts tagged with "API"

v2.2.0 is here 🧭​

A minor release with two behaviour changes worth flagging upfront, an 8-day retention window on generated artworks, two fixes, and a top-to-bottom refresh of these docs.

What's changed​

nsfw flag now consistent across models ⚖️​

Every artwork endpoint now follows the same rule: nsfw: true means allow NSFW output, and nsfw: false (the default) suppresses it. Selena and Selena Variete used to be inverted, so if you were sending nsfw: true to those models expecting suppression, flip it to nsfw: false.

Default promptMode flipped to servitor 🪄​

The default promptMode changed from seer (AI prompt enhancement) to servitor (raw prompt) across Selena, Selena Variete, Hamilton, and Quantum. To keep prompt enhancement, pass promptMode: "seer" explicitly.

8-day retention on generated artworks 🗓️​

Artwork records, their input images, and their generated images are all removed 8 days after creation. After the window, GET /v2/artworks/:id returns 404 for that ID. If you need to keep an image around long-term, download it within those 8 days.

Fixes 🔧​

• GET / no longer returns an empty route list. Under NestJS 11 and Express 5, the metadata endpoint's internal router lookup broke silently. Restored, so the routes field is populated again.
• Selena Variete now honours promptMode. The field used to be ignored.
• NSFW suppression now applies cleanly on Selena and Selena Variete. It was previously being applied twice.

Documentation rework 📚​

We've also taken the docs through a top-to-bottom pass:

• A new Introduction page that gives you a real overview of what the API does (image generation, image editing, image utilities, and the Textcraft LLM) and how to authenticate.
• Section order is now stable and meaningful: General, Artworks, Selena, Quantum, Hamilton, Photo Pilot, Textcraft, Style DNA, Organizations.
• Every endpoint summary and description has been rewritten in plain language so you can tell what a route does at a glance.
• The version selector locks to V2 (current) and V1 (deprecated), with a deprecation banner across every V1 page.
• Internal-only endpoints no longer appear in the docs or in the GET / route list, keeping the public surface tight.

Getting started​

If you're on v2.1.0, the only changes you might feel at the wire level are the two behaviour flips above (nsfw semantics, promptMode default). Everything else is additive or transparent.

For the full reference, see our API documentation.

Happy Creating! 🎨✨
The Tengr.ai Team

v2.1.0 is here ✨​

A week after v2.0.0, this is a focused follow-up. The big things are truly non-blocking artwork creation, a model selector on Face Refiner, structured error shapes, the Infinity styles working in image-to-image, and a stack of security upgrades.

What's new​

Truly non-blocking artwork creation ⚡​

Every v2 artwork-creation endpoint backed by the in-house pipeline (selena, selena-variete, hamilton, photo-pilot, face-refiner, face-swap, segmentation, depth-map, detailer-upscaler, quantum, remove-background) used to return 202 only after prompt building and the upstream dispatch finished. Both can take long enough to overrun client HTTP timeouts.

Now the artwork is persisted with status: STARTING and the 202 returns in milliseconds. Prompt building (which can chain 3-4 LLM calls) and the upstream dispatch run in the background. If anything fails after the 202, the artwork lands in FAILED and the failure flows out through your webhook, not a synchronous 5xx.

Zod validation errors still return 4xx synchronously, same as before.

Face Refiner accepts a model selection 👤​

POST /v2/artworks/face-refiner now takes an optional model parameter. Leave it off and the request fans out to all available Tengr.ai face-refining models (the default behaviour). Pass a specific model name and the request runs only on that one. This brings Face Refiner in line with how Selena and Selena Variete handle model selection.

Structured error responses 📋​

Every v2 endpoint now returns errors in a consistent { status, message } shape, both on GET /v2/artworks/:id and on outbound webhooks. The v1 wire shape is frozen, so older integrations see the same errors they always did.

Infinity styles in image-to-image 🌀​

The seamless Infinity styles (infiny-weave, infiny-pattern, infiny-flow) now work in Selena Variete, not just in text-to-image. Useful when you want tileable backgrounds or repeating patterns starting from an existing image.

Hamilton style tuning 🎚️​

A handful of Hamilton styles got their default colour grade softened, so generations look less heavily stylised out of the box.

Security updates 🛡️​

A round of dependency upgrades to keep the stack current and clear the latest security advisories. No action needed on your end.

Performance and reliability​

• Better CPU utilisation on multi-core machines via optional worker processes.
• A self-healing job that picks up any artwork whose status update or webhook didn't arrive and finishes it correctly, so no generation silently stays stuck.
• A unified error vocabulary across all generation models, so the same kind of upstream failure looks the same from your side regardless of which model produced it.

Getting started​

If you're already on v2.0.0, this release is backward compatible. The async improvements speed things up on their own. The new options are opt-in.

For request examples and migration notes, see our API documentation.

Happy Creating! 🎨✨
The Tengr.ai Team

v2.0.0 is here 🚀​

v2 is a major rework of the public API. We've split the surface into two versions side by side: a frozen v1 (the v1.4.0 set, on Supabase JWT auth) and a new v2 surface on API-key auth with two-dimensional access control, async generation with HMAC-signed webhooks, and a wider catalogue of in-house generation endpoints. Existing v1 integrations keep working unchanged.

What's new​

Versioning split 🪟​

Every route that shipped at v1.4.0 keeps its /v1/... form. The same routes also exist at /v2/... under API-key auth. Anything we introduced after v1.4.0 (Quantum, Selena Variete, Face Refiner, Face Swap, Segmentation, Depth Map, Detailer Upscaler) is v2-only. Migrate when it suits you.

API key auth with scope and domain 🔐​

v2 keys carry two dimensions. Scope is a strict hierarchy: admin > manager > generator. Domains are the feature areas the key can reach: artworks, style-dna, textcraft, or * for all of them. Both have to pass on every request.

Org admins manage keys through POST /v2/organizations/:orgId/api-keys (create) and DELETE /v2/organizations/:orgId/api-keys/:keyId (revoke). The full key is returned once at creation, so store it somewhere safe.

New in-house generation endpoints 🎨​

• Quantum for text-to-image with high coherence and detail, plus its own styles and aspect ratios.
• Selena Variete for image-to-image variations, with a catalogue of processing modes.
• Face Refiner for cleaning up portrait detail.
• Face Swap for replacing a face from a source image onto a target.
• Segmentation for masks from prompts or points.
• Depth Map for grayscale depth output.
• Detailer Upscaler for sharpening fine details with optional upscaling.
• Upscaler at /v2/artworks/upscaling with the same Selena-backed upscaler.

Async workflows with signed webhooks 📤​

Every v2 generation endpoint returns 202 Accepted immediately with an artwork ID. The final result lands at the webhook URL you supplied with the request, signed with HMAC-SHA256. Each delivery carries Tengrai-Timestamp and Tengrai-Signature headers, computed over ${timestamp}.${rawBody} using your organization's webhook secret. You can rotate the secret through POST /v2/organizations/:orgId/webhook-secret.

v1 endpoints stay synchronous and unsigned, so existing integrations don't have to change.

Style DNA goes async in v2​

The image-producing Style DNA endpoints under /v2/style-dna/* now return 202 and complete via webhook. Specifically: generate, generate/asset, edit, edit/style-aware, remove-background, and generate-anchor. The two sub-second LLM-only endpoints (generate-style-options and refine-prompt) stay synchronous.

For generate/asset with backgroundMode: transparent, the server chains generation and background-removal automatically. Your webhook fires once, on the final result.

Getting started​

The v2 reference includes request examples, webhook setup, and the auth flow. Head to our API documentation for the full v2 surface.

If you're on v1, your existing integrations keep working at /v1/.... Migrate when the new endpoints, webhooks, or auth model fit your needs.

Happy Creating! 🎨✨
The Tengr.ai Team

🎉 Announcing Tengr.ai API v1.4.0 – Introducing Hamilton & Photo Pilot! 🎨✨​

We're excited to announce Tengr.ai API v1.4.0 🚀, a milestone release featuring powerful new image generation models and completing our Style DNA feature set. This update brings professional-grade tools for photographers, designers, and creative businesses.

🎯 What's New in Tengr.ai API v1.4.0?​

🔑 New AI Models for v1 🤖​

Hamilton – High-resolution image generation with exceptional prompt fidelity:

• POST /v1/artworks/hamilton – Generate stunning high-resolution artwork with strong adherence to your prompts
• GET /v1/artworks/hamilton/aspect-ratios – Discover supported aspect ratios
• GET /v1/artworks/hamilton/styles – Explore available artistic styles

Photo Pilot – Intelligent photo editing and enhancement:

• POST /v1/artworks/photo-pilot – Edit photos with surgical precision while preserving original concepts and feel
• GET /v1/artworks/photo-pilot/aspect-ratios – View supported dimensions

🔑 Utility Endpoints 🛠️​

Background Removal:

• POST /v1/artworks/remove-background – Remove image backgrounds with precision. Pair with Style DNA for branded replacements!

🔑 Expanded Style DNA for All Models 🎨​

Style DNA endpoints now available across Selena, Hamilton, and Photo Pilot models:

• GET /v1/artworks/selena/styles – Browse Selena's artistic palette
• GET /v1/artworks/selena/aspect-ratios – Selena-specific dimensions

🔑 Enhanced API Key Management​

Organizations can now rotate and manage webhook secrets:

• POST /v1/organizations/{orgId}/webhook-secret – Generate or rotate your organization's webhook signing secret for secure event delivery

🎨 What This Means for Your Business​

With v1.4.0, you now have a complete toolkit:

• Selena for creative, stylized imagery 🎨
• Hamilton for high-fidelity, photorealistic outputs 📸
• Photo Pilot for intelligent photo enhancement ✨
• Style DNA for consistent, branded visual control 🎯
• Background Removal for seamless integration 🎪

Perfect for agencies, e-commerce platforms, content creators, and enterprises building visual-first applications.

💻 Ready to Get Started?​

For more detailed API documentation 📚, including the complete Style DNA workflows, model comparisons, and usage examples, please visit our API documentation.

This is a significant step forward for Tengr.ai's v1 API. We're excited to see what you'll build! 🚀

Happy Creating! 🎨✨
The Tengr.ai Team

🎉 Announcing Tengr.ai API v1.3.0 – Meet Style DNA! 🎨✨​

We're thrilled to announce Tengr.ai API v1.3.0 🚀, introducing Style DNA — our breakthrough feature for style-driven image generation and editing. This release empowers business users with unprecedented control over artistic direction and brand consistency in AI-generated content.

🎯 What's New in Tengr.ai API v1.3.0?​

🔑 Style DNA: Your Artistic Control, Amplified 🎨​

Style DNA is a new domain in Tengr.ai that lets you generate, edit, and manage images with consistent stylistic direction. Whether you're building branded content, maintaining visual consistency, or exploring creative variations, Style DNA gives you the tools.

Core Style DNA Endpoints:

• POST /v1/style-dna/generate – Generate images with a specific style 🎨
• POST /v1/style-dna/edit – Edit existing images with style guidance 🖌️
• POST /v1/style-dna/edit/style-aware – Modify images while preserving their original style
• POST /v1/style-dna/remove-background – Remove backgrounds from images with style preservation
• POST /v1/style-dna/generate/asset – Generate styled asset backgrounds 📦

🔑 API Key Management for Organizations 🔐​

Introducing role-based API key scoping for enterprise teams:

• POST /v1/organizations/{orgId}/api-keys – Create scoped API keys with role-based permissions:

  • Admin – Full access to all operations
  • Manager – Management and generator-level operations
  • Generator – Generation operations only

• DELETE /v1/organizations/{orgId}/api-keys/{keyId} – Revoke keys as needed

🔑 Manager & Generator Role Endpoints​

For Managers:

• POST /v1/style-dna/generate-style-options – Generate 3–5 style variant options for exploration
• POST /v1/style-dna/refine-prompt – Refine prompts based on feedback
• POST /v1/style-dna/generate-anchor – Generate anchor style images from text

For Generators:

• All Style DNA generation and editing endpoints
• All standard artwork creation endpoints

🚀 Enhanced Developer Experience​

This release lays the foundation for team-based workflows, letting you implement fine-grained access control across your organization. Perfect for agencies, creative studios, and enterprises managing multiple projects and team members.

💻 Ready to Get Started?​

For more detailed API documentation 📚, including request examples, style DNA workflows, and usage tips, please visit our API documentation.

We can't wait to see how Style DNA will transform your creative applications! 🎨✨

Happy Creating! 🎨✨
The Tengr.ai Team

🎉 Announcing Tengr.ai API v1.2.0 – Enhanced Image Generation Control! 🎨✨​

We're thrilled to introduce Tengr.ai API v1.2.0 🚀, packed with improvements to give you more fine-grained control over your AI-generated artwork. This release responds directly to business user feedback, providing greater flexibility in image dimensions and transparent backgrounds.

🎯 What's New in Tengr.ai API v1.2.0?​

🔑 Key Updates​

• Unified Selena Endpoint: The POST /v1/artworks/txt2img endpoint is now POST /v1/artworks/selena, reflecting our focus on naming endpoints after the AI models they use. This aligns with future releases and makes the API structure more intuitive.

• Transparent Background Support 🎨:
  Added the transparentBackground field to the Selena artwork creation request. When enabled, generated images will have a transparent background instead of the default opaque background, giving you more creative flexibility for integration into your applications.

• Improved Upscaling Sync ⬆️:
  Enhanced sync between selena and upscaling artwork task types to ensure consistent behavior and support for our latest generation capabilities.

💻 Ready to Get Started?​

For more detailed API documentation 📚, including request examples and usage tips, please visit our API documentation.

We're excited to see how these enhancements will help you create even more stunning AI-generated artwork! 🎨✨

Happy Creating! 🎨✨
The Tengr.ai Team

🎉 Announcing Tengr.ai API v1.1.0 for Business Users – Enhance Your Text-to-Image Experience! 🎨✨​

We’re excited to introduce Tengr.ai API v1.1.0 🚀, the latest update tailored for business users who want to elevate their creative projects with enhanced AI-powered features. This version builds upon our powerful text-to-image (txt2img) capabilities, making it easier and more versatile to generate stunning visuals from text prompts.

🎯 What's New in Tengr.ai API v1.1.0?​

In this release, we’ve added new features and improvements to streamline your workflow and give you greater control over your artwork generation. Here’s what’s new:

🔑 New Features​

• New API Endpoint:

  • POST /v1/artworks/upscaling – Introducing a dedicated endpoint for upscaling existing artworks or other images, allowing you to enhance the quality and resolution of your generated images with ease. This feature is perfect for refining visuals for high-quality applications.

• Enhanced POST /v1/artworks/txt2img Request:

  • Added width and height fields in the request body, which are now mutually exclusive with the aspectRatio field. This provides you with the flexibility to specify exact dimensions for your images or choose a predefined aspect ratio, depending on your project needs.

🎨 More to Come 🚀​

Tengr.ai API v1.1.0 marks another step forward in our journey 🌟. We’re continually working on exciting new features 🛠️ and improvements that will expand our platform’s capabilities. Stay tuned for more updates that will introduce further customization options ⚙️, advanced AI models 🤖, and deeper integrations to enhance your creative projects.

💻 Ready to Get Started?​

For more detailed API documentation 📚, including the latest features, request examples, and usage tips, please visit our API documentation.

We can’t wait to see how you’ll leverage the new features in Tengr.ai to create amazing content! 🚀

Happy Creating! 🎨✨
The Tengr.ai Team

🎉 Announcing Tengr.ai API v1.0.0 for Business Users – Unlock the Power of Text-to-Image Creativity! 🎨✨​

We’re thrilled to announce the release of Tengr.ai API v1.0.0 🚀, specifically designed for business users looking to supercharge their projects with AI-powered text-to-image (txt2img) generation. With this release, developers, creators, and innovators in businesses of all sizes can seamlessly integrate the next generation of creative tools into their workflows, transforming text prompts into high-quality, AI-generated visuals in just seconds.

🎯 What's in Tengr.ai API v1.0.0?​

In this first version, we’ve packed the core features you need to start generating custom AI-powered artwork at scale 🎨, with robust API endpoints and flexible customization options. Whether you’re building an app, enhancing a design workflow, or creating art for marketing or branding projects, Tengr.ai makes it possible with just a few simple API calls.

🔑 Key Endpoints​

Here’s an overview of the available endpoints:

• GET / – Fetch basic API metadata 📊, including version and current endpoints information.
• GET /health – Check the health status of the API ✅.
• GET /v1/models – Explore available AI models 🧠 you can use for generating images.
• GET /v1/aspect-ratios – Retrieve a list of supported aspect ratios 📐 for your image outputs.
• GET /v1/styles – Discover the artistic styles 🎨 you can apply to your text prompts.
• GET /v1/artworks – List previously generated artworks 🖼️.
• GET /v1/artworks/:id – Retrieve detailed information about a specific artwork by its ID 🔍.
• POST /v1/artworks/txt2img – The heart of the API ❤️! Submit a text prompt 📝 and get an AI-generated image in response 🖼️. It’s as easy as sending a request with your desired text input and letting our models do the magic 🧙‍♂️.
• DELETE /v1/artworks/:id – Manage your creations 🗑️ by deleting artwork when needed.

🚦 Built-In Rate Limiting​

We’ve implemented rate limiting to ensure fair usage ⚖️ and API stability for all users. If your business requires higher throughput or custom rate limits, don’t hesitate to reach out to us at support@tengrai.com. We’re here to help your project scale 🌱.

🔥 Why Tengr.ai?​

Our vision is simple: to empower businesses with cutting-edge AI tools 🛠️ that unlock new creative possibilities ✨. Whether you’re aiming to generate stunning visuals for your product marketing 📈, create custom imagery for branding 🎨, or drive innovation in creative applications, Tengr.ai provides the power to bring these concepts to life through AI-powered art generation.

With the txt2img feature, the possibilities are endless 🔄. Input a description, specify a style 🖌️, and watch as our AI transforms your words into breathtaking visuals 🎇. This is the future of creative automation, and we’re excited to partner with you on this journey 🛤️.

🎨 More to Come 🚀​

Tengr.ai API v1.0.0 is just the beginning 🌟. We’re already working on exciting new features 🛠️ and improvements to expand the creative and functional capabilities of our platform. Stay tuned for updates that will introduce more customization ⚙️, advanced AI models 🤖, and deeper integrations to make Tengr.ai even more powerful for business users 💼.

💻 Ready to Get Started?​

For more detailed API documentation 📚, including request examples and usage tips, please visit our API documentation.

We can’t wait to see how your business will create with Tengr.ai! 🚀

Happy Creating! 🎨✨
The Tengr.ai Team