> For the complete documentation index, see [llms.txt](https://docs.saas.haut.ai/haut.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.saas.haut.ai/haut.ai/releases/changelog/version-2026-23.md).

# Version 2026-23

## Release date

June 3, 2026

## Face Skin Analysis 3.0

Face Skin Analysis 3.0 delivers improved accuracy across lines-related metrics, reduced model latency, a faster skin-mask filtering step, and a new customer-facing parameter, **Forehead Lines** (**Lip Lines** was announced in this release but is not yet returned by the API).

### Lines algorithm update

The lines algorithm has been rewritten with a unified model architecture and an extended set of customer-facing parameters.

* **Forehead Lines** detects horizontal forehead wrinkles and returns grade, fine/deep classification, vector masks, line amount, and relative length. Forehead Lines uses grade values from 1 to 6. Learn more: [Forehead Lines](/haut.ai/haut.ai-features/algorithms/face-skin-analysis-3-0/lines/forehead-lines.md) and [Forehead Lines Parameter Result Scheme](/haut.ai/developers/saas-api-overview/api-for-face-skin-analysis-3.0/lines/forehead-lines.md).
* **Lip Lines** detects wrinkles that radiate outward from the mouth and encircle the lips, returning grade, fine/deep classification, vector masks, line amount, and relative length. Lip Lines uses grade values from 1 to 5. Lip Lines is not yet available through the API; documentation will be published when it ships.
* **Improved perioral-region detection** for lip lines, nasolabial wrinkles, marionette lines, jowls, and sagging. These metrics are now driven by a unified perioral model.
* **Improved forehead-region detection** for forehead lines and glabellar lines. Glabellar lines now work with all three sides and use aligned classification and segmentation.
* **Glabellar lines length calculation updated**. The method used to compute glabellar line length has changed slightly, so the reported length for the same image may differ between the previous version and this release.
* **Reduced model latency** for lines-related processing.

### Score drift and compatibility

Existing line parameters (`lines`, `fine_lines`, `deep_lines`, `under_eye_lines`, `crows_feet_lines`, `nasolabial_wrinkles`, `marionette_lines`, `jowls`, `sagging`, `eyes_lines`) keep the same output field names. New top-level entries are added for Forehead Lines, Lip Lines, and their fine/deep variants.

One-time prediction shifts are expected on perioral- and forehead-region metrics after upgrade. This is a step change to a new baseline, not a regression. Most affected results are expected to shift by approximately 1-3 points; larger shifts can occur for images where the previous model performed poorly.

| Score        |  MAE | Median | 75th percentile | 90th percentile | Max drift |
| ------------ | ---: | -----: | --------------: | --------------: | --------: |
| `lines`      | 2.45 |      1 |               3 |               7 |        26 |
| `fine_lines` |  3.0 |      1 |               2 |              11 |        37 |
| `deep_lines` |  1.8 |      0 |               1 |               6 |        39 |

### Skin segmentation speed improvement

The skin-mask filtering step has been optimized, noticeably speeding up the face analysis pipeline on high-resolution images.

* **Minor mask boundary change**: masks may have an approximately 1-pixel tighter fringe than previous masks, with at least 99.99% IoU.
* **Small one-time score shifts** can occur on metrics sensitive to the mask edge, primarily per-area pigmentation and forehead lines. This is not a regression, and no customer action is required.

## Platform defaults

New companies now start with the current application defaults instead of legacy analysis defaults.

* Face Skin Metrics 2.0 and Face Skin Metrics 2.0-based Recommendation System 2.0 defaults are no longer enabled for new companies by default. New setups should use Face Skin Analysis 3.0 and the current Recommendation System 2.0 flow. Learn more: [Face Skin Analysis 3.0](/haut.ai/haut.ai-features/algorithms/face-skin-analysis-3-0.md) and [Product Recommendation System 2.0](/haut.ai/haut.ai-features/product-recommendation-system-2.0.md).
* Hair Metrics 2.0 is no longer enabled for new companies by default. Existing companies are not changed by this release. Learn more: [API for Hair Metrics 2.0](/haut.ai/developers/saas-api-overview/api-for-hair-metrics-2.0.md).

## Consumer Apps

### Navigation and settings

* The SaaS menu section is now named **Consumer Apps**. Existing apps and integration URLs are unchanged. Learn more: [Consumer Apps](/haut.ai/haut.ai-features/consumer-apps.md).
* Consumer Apps can now hide the Favorites feature when it is not part of the desired end-user journey.
* Selected hair parameter results can now be hidden in Consumer App results, giving brands more control over which hair analysis outputs are shown to end users. Learn more: [Configure Advanced Settings](/haut.ai/haut.ai-features/consumer-apps/configure-advanced-settings.md).

### Hair recommendations

Hair recommendation logic now handles survey and profile inputs more consistently.

* If the hair procedures/damage question is not included in the Consumer App survey, the recommendation engine ignores the hair-condition filter instead of excluding otherwise relevant products.
* Secondary hair concerns are capped at two.
* User-provided curliness from the survey now takes priority over algorithm-detected curliness for recommendation matching.

Learn more: [Hair Recommendation Algorithm](/haut.ai/haut.ai-features/product-recommendation-system-2.0/recommendation-algorithm/hair-recommendation-algorithm.md) and [Hair User Tags](/haut.ai/haut.ai-features/product-recommendation-system-2.0/user-tags/hair-user-tags.md).

### Analytics and Klaviyo

* The Klaviyo **Skin Analysis Complete** event now includes `batch_id`, making it easier to connect CRM events with a specific analysis session. Learn more: [Data Sent to Klaviyo](/haut.ai/quick-start-guides/no-code-app-for-e-commerce-businesses/integrate-consumer-app-with-klaviyo-crm/data-sent-to-klaviyo.md).
* For companies with Google Analytics enabled, Consumer App analytics events are now always forwarded to the parent page via `postMessage`. Learn more: [Send Consumer App Analytics Events to Third-Party Analytics](/haut.ai/haut.ai-features/consumer-apps/connect-google-analytics-to-your-consumer-app/analytics-event-forwarding-via-postmessage.md).
* Forwarded analytics events no longer include `subject_id` or raw skin-analysis results. If your website uses custom parent-page listeners that depended on those fields, update the listener to use the supported analytics event payload.

## API keys

API key management now includes stronger token handling and easier key identification.

* Private tokens now use opaque bearer tokens with hash-based lookup for improved security. Existing valid tokens continue to work.
* The API Keys table now shows the last 4 characters of each key, making it easier to identify keys without exposing the full token.
* API keys can now be created without an expiration date. We still recommend setting an expiration date and rotating keys regularly when your integration allows it. Learn more: [Authentication and Token Management](/haut.ai/developers/authentication-and-token-management.md).

## Datasets, recommendations, and reports

* In Datasets, clicking a subject now opens the subject's photo list instead of opening the report directly. Learn more: [Subjects](/haut.ai/haut.ai-features/datasets/subjects.md).
* PDF reports now categorize products by concern in the **You may also like** block, making alternative product recommendations easier to review. Learn more: [PDF reports](/haut.ai/haut.ai-features/datasets/get-a-report/pdf-reports.md) and [Recommendations View in Consumer App](/haut.ai/haut.ai-features/product-recommendation-system-2.0/recommendations-view-in-consumer-app.md).

***

{% hint style="warning" %}
Not a Haut.AI client yet?
{% endhint %}

<a href="https://haut.ai/book-a-demo?utm_source=referral&#x26;utm_medium=saas_docs&#x26;utm_campaign=Docs_SaaS_HautAI" class="button primary" data-icon="user-vneck">Book a demo</a> <a href="https://haut.ai/product/ai-skin-analysis?utm_source=referral&#x26;utm_medium=saas_docs&#x26;utm_campaign=Docs_SaaS_HautAI" class="button secondary" data-icon="external-link">Platform overview on haut.ai</a>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.saas.haut.ai/haut.ai/releases/changelog/version-2026-23.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.