3.4.0

Note that this release bumps the WordPress minimum from 6.6 to 6.7.

Added

• New Content Generation Feature, allowing the creation of draft content from a summary (props @dkotter, @jeffpaul, @Sidsector9, @fabiankaegy via #859, #864).
• New Recommended Content block powered by the OpenAI Embeddings Provider (props @dkotter, @fabiankaegy, @jeffpaul, @Sidsector9 via #881).
• Default template global settings for Recommended Content Block (props @Sidsector9, @dkotter via #917).
• Support for the new OpenAI gpt-image-1 image generation model (props @dkotter, @jeffpaul, @Sidsector9 via #897).
• Support for WooCommerce product-based title and excerpt generation (props @faisal-alvi, @dkotter, @jeffpaul via #871).
• New Processing Mode setting to our Image Processing Features, allowing you to only have those Features trigger when someone manually requests it, rather than automatically on upload (props @dkotter, @jeffpaul, @Sidsector9 via #862).
• Support for selecting new OpenAI Audio Transcription models, GPT-4o mini Transcribe and GPT-4o Transcribe (props @dkotter, @jeffpaul, @iamdharmesh via #875).
• Support for selecting the new GPT-4o mini TTS model in our Text to Speech Feature (props @dkotter, @iamdharmesh via #876).

Changed

• Switch to the Gemini 2.5 Flash model from the Gemini Pro model (props @dkotter, @MadtownLems, @jeffpaul, @iamdharmesh via #927).
• Bump WordPress minimum from 6.6 to 6.7 (props @faisal-alvi, @dkotter, @jeffpaul via #871).
• Bump WordPress "tested up to" version to 6.8 (props @sksaju, @dkotter via #900).
• Use the attached release asset in our plugin updater (props @dkotter, @jeffpaul via #870).
• Set the GPT-4o mini TTS model as our new default in our Text to Speech Feature (props @dkotter, @iamdharmesh via #876).
• Minor text tweaks to our registration language (props @dkotter, @jeffpaul, @Sidsector9 via #868).
• Don't show the registration admin notice when the welcome guide is showing (props @dkotter, @jeffpaul, @Sidsector9 via #868).
• Don't disable the Settings button if the Feature is turned off (props @dkotter, @faisal-alvi via #867).
• Update the layout of the Generate Titles modal (props @dkotter, @iamdharmesh via #860).

Fixed

• Fatal error where $post was null (props @sksaju, @dkotter via #905).
• Ensure we properly store the post type setting to avoid PHP notices (props @JiveDig, @dkotter, @Sidsector9 via #893).
• Fix condition to prevent warnings when $args['args'] is not defined (props @dilipbheda, @dkotter via #896).
• Ensure that the Text-to-Speech feature options are available only for the allowed post types (props @iamdharmesh, @faisal-alvi, @dkotter via #878).
• Ensure the embeddings update notice doesn't show if it isn't needed (props @dkotter, @jeffpaul, @Sidsector9 via #868).
• Fix various Gutenberg deprecation notices added in WordPress 6.5 and 6.6 (props @dkotter, @iamdharmesh via #860).
• Ensure we pass the most recent version of the content in our title and excerpt generation Features (props @dkotter, @iamdharmesh via #860).

Security

• Bump tar-fs from 3.0.8 to 3.0.9 (props @dependabot[bot], @dkotter via #929).
• Bump react-router from 7.1.5 to 7.5.2 and react-router-dom from 7.1.5 to 7.5.2 (props @dependabot[bot], @dkotter via #899).
• Bump http-proxy-middleware from 2.0.6 to 2.0.9 (props @dependabot[bot], @dkotter via #894).
• Bump axios from 1.7.9 to 1.8.3 (props @dependabot[bot], @dkotter via #869).

Developer

• Rename the DallE Provider class to Images. If you directly extend that class yourself, you'll need to update your code to account for this. Also updated a handful of other references to DALLE to Images (props @dkotter, @jeffpaul, @Sidsector9 via #897).
• Rename the OpenAI Whisper Provider to SpeechToText. If anyone extended the Whisper class, you'll need to update your code to account for this (props @dkotter, @jeffpaul, @iamdharmesh via #875).
• New filter, classifai_googleai_gemini_api_model, making it easier to change the model Google AI uses (props @dkotter, @MadtownLems, @jeffpaul, @iamdharmesh via #927).
• New filters, classifai_openai_speech_to_text_model and classifai_openai_speech_to_text_api_url, around the Speech to Text model and API URL, allowing those to easily be changed (props @dkotter, @jeffpaul, @iamdharmesh via #875).
• New filters, classifai_openai_text_to_speech_api_url and classifai_openai_text_to_speech_model, Text to Speech model and API URL, allowing those to easily be changed (props @dkotter, @iamdharmesh via #876).
• Initial set of data flow diagrams (props @christianc1, @dkotter via #902).
• New developer doc that outlines which Providers we support and which Features each Provider supports (props @dkotter, @jeffpaul via #882).
• New developer doc that outlines how data is used within ClassifAI (props @dkotter, @jeffpaul via #874).
• Ensure the final release asset gets attached properly to the release (props @dkotter, @jeffpaul via #870).
• Update release instructions (props @jeffpaul, @dkotter via #903).
• Update @10up/cypress-wp-utils from 0.4.0 to 0.5.0 (props @dkotter, @jeffpaul via #872).
• Update all third-party actions our workflows rely on to use versions based on specific commit hashes (props @dkotter, @jeffpaul via #873).

New Contributors

• @fabiankaegy made their first contribution in #864
• @JiveDig made their first contribution in #893
• @dilipbheda made their first contribution in #896
• @MadtownLems made their first contribution in #927

---

Full Changelog: 3.3.1...3.4.0
View closed items in the milestone.