Hello 
I’m excited to be posting our API’s first release thread!
This release includes more transparent pricing, five new models, as well as makes our Bloom and Starlight models available to all customers (not just enterprise-exclusive) . We’ve also revamped our Developer Documentation as well as updated our API Playground, and Credit Calculator to reflect our recently released models.
Pricing Update
We’ve simplified how API pricing works as well as cut prices a bit for a few models.
Note: as part of this change, we have not increased the price of our models.
The biggest change is structural. Previously, every model had its own credit cost that could vary depending on the job. We’ve reorganized models into “families” and standardized how pricing scales.
Below is an overview of our upscaling families for image and video.
Image
Image models are now priced based on output megapixels per credit.
There is now one rate per model family, and it scales linearly with output size. No more guessing what a job will cost — if you know the output resolution, you know the price.
For example, our “Gigapixel’ model family is priced at 24 MP per Credit. So for every model in that family, each credit spent gets you up to 24 MP of output before you are charged an additional credit.
Video
Video models use published pricing tables for quick references at common output resolutions and video durations.
Each model family has standardized pricing with a few exceptions. For example, the “Fast” sub-tiers of a given model family is roughly half the price of the family’s price.
The updated Credit Calculator reflects all of these changes, so you can plug in your expected workloads and see exactly what things cost before committing!
Un-Gating Premium Models
Bloom and Starlight models were previously available only to enterprise API customers. Starting today, we’ve made these models available to everyone, for the same price—a huge step in ensuring that Topaz users are always getting the latest and greatest of our models!
Here are all “Premium” models that you now have access to:
-
Bloom Creative — Creative upscaling for AI generated images and artwork
-
Bloom Realism — Realistic and natural detail enhancement for AI-generated images
-
Astra Creative — Transform GenAI video to 4K with creative upscaling that generates dynamic new texture and detail
-
Starlight Precise 1 — Restore footage once considered impossible with fully automatic diffusion enhancement
-
Starlight Precise 2 — Bring realism to GenAI video with natural faces, skin tones, and stable motion
-
Starlight HQ — Maximum quality generative enhancement for modern high-resolution video
-
Starlight Sharp — Fast local restoration with sharper detail for clean, progressive video
-
Starlight Mini — Diffusion restoration for archival, degraded, and interlaced footage
-
Starlight Fast 1 — First generation generative upscaling for speed
-
Starlight Fast 2 — 2x faster diffusion upscaling with higher fidelity
New Models
Five new models are now available in the API:
-
Starlight Fast 2 — 2x faster diffusion upscaling with higher fidelity
-
* Starlight Precise 2.5 — Improved realism for AI-generated video with enhanced faces, texture detail, and text clarity
-
Starlight HQ — Max quality generative enhancement for modern high-resolution video
-
Background Removal (Image) — Remove backgrounds from still images
-
Gaia 2 — Next-generation frame-by-frame 2× upscaling for animation
Reference the linked model pages to view full documentation that includes descriptions, supported endpoints and parameters, as well as output examples.
Documentation, Playground, and Calculator
The API docs have been updated to reflect the new model families and pricing structure. We’ve also added several new pages covering our new categorization of models as well as how to choose the best model for your specific use case.
The playground and calculator also now reflect all available models so that you can test out and estimate costs for any combination of models and resolutions.
We know that our API docs, playground, and calculator are important resources for all of our customers, so we’d love to hear your feedback, questions, and ideas! If you’d like to get involved, feel free to leave comments here in our Forum or in our Developer Discord!
Going forward, we will continue to iterate on the docs, pricing, and more in order to make the API models as as easy to understand and use as possible!
Changelog
-
Massive pricing simplification / implemented model families
-
Large Documentation Update
- Overall revamp, included new pricing simplification / family model structure
-
Launched five new models
-
Starlight Fast 2
-
Starlight Precise 2.5
-
Starlight HQ
-
Background Removal (Image)
-
Gaia 2
-
-
Un-Gated “Premium” Bloom & Starlight models
-
Bloom Creative
-
Bloom Realism
-
Astra Creative
-
Starlight Precise 1
-
Starlight Precise 2
-
Starlight HQ
-
Starlight Sharp
-
Starlight Mini
-
Starlight Fast 1
-
Starlight Fast 2
-
-
Updated API Credit Calculator to include all available models
-
Updated API Playground to include all available models
For any issues, reach out to support@topazlabs.com. Enterprise inquiries can be directed to enterprise@topazlabs.com.
Aiden Warnke
Product Manager
Enterprise & API
