What's new in Gibrid

1October 16, 2026 — Studio, voice and music, chat, invites

Today the school gained guides for all the new features. In short, what is new and where to find it.

Studio.

• Animate photo and “Animate selected”: a card in an empty feed with the price on the button; in a batch, up to 10 frames with one motion.
• Chains: “Product → 5 creatives” and “Portrait → photoshoot → video”; the price of the whole chain is shown before you start.
• Ad from a product link: a product page, 5 frames from a script, and a required tick confirming you own the rights to the photo.
• Translate a video: 9 languages, a catalog voice or your own clone, translated subtitles; lips are not re-synced.
• Subtitles, quick image edits and “Edit the clip with words”.
• Other format, brand plate, montage and archive: free tools for finishing a clip.
• Compare models, a public link to a work, the Done alert and composer extras: dictation, price before launch.

Voice and music.

• Voiceover: the Fast / Natural / Max quality, fine tuning, dialogue, voice comparison, My recordings.
• Voice library: thousands of ElevenLabs voices, Pro plan or above, or a balance of $1+.
• Music on Suno V6 and what to do with a finished track: extend, cover, replace a section, stems, WAV, timed lyrics, video.

Chat and account.

• Auto, the “≈ N G” price chip, Continue and compression of long chats; temporary chat and the “Do not remember” switch.
• Research, big files, message search and voice talk on the web.
• The send-confirmation card: an email or message goes out only after your Send.
• Invites with rewards and the new Settings.

2October 15, 2026 — School: courses now cover the newest model versions

• Seedance 2.5. The “Seedance” course is now about Seedance 2.5: up to 30 seconds, up to 30 images, 10 videos and 10 audio clips as references, a 480p draft and a 1080p finalize with Seedance 2.5 Finalize. Seedance 2.0 is still in the Studio — see “Older versions” in the course.
• Wan 3.0, MiniMax H3, HappyHorse 1.1. The “Wan”, “Hailuo” and “HappyHorse” courses now cover Wan 3.0 (and Prime), MiniMax H3 together with Hailuo 2.3, and HappyHorse 1.1.
• Images. Courses on Nano Banana 2.1, FLUX 2, Seedream 5.0 and GPT Image 2.5. Previous versions moved to an “Older versions” block.
• Grok Imagine 1.5. The course now shows the correct duration (1–15 seconds) and quality (up to 1080p).
• Model catalog. Older versions have a “Previous version” banner on their model pages linking to the newest. Pages for the newest chat models were added: Claude Opus 5.5, Sonnet 5.5, Haiku 5.5, Fable 5.1, GPT-6.1 Sol, GPT-6 Astra and Luna, Grok 4.7, Qwen 3.8 Max Prime, DeepSeek V4.1 Flash, Muse Spark 1.3, Mistral Large 4.

314–15 October 2026: Studio, storyboards and voice

• Up to 10 frames. In the storyboard “Settings”, the “Frames” field accepts values from 3 to 10. The main screen assembles a whole video from several clips (details);
• Simple mode. Turn off the Advanced mode switch to get a short path: “pick a frame → paste text → pick a voice → one Voice the frame button”. Which screen opens depends on the “Advanced mode” switch (on — the regular screen, off — the simple one); your choice is remembered in the browser;
• Consistent scene option. A Keep the scene consistent switch has appeared in “Settings”: the first frame is drawn first, and the rest take the place, lighting and clothing from it but are shot from new angles. It is off by default and is handy for a video set in one location;
• Your own animation model. Next to the Animate all button there is an “Animation model” field, and for an individual frame it is Frame settings → “Animation model”. Prices in ⚡ are shown in the model list;
• An idea is not a chat. In the idea field, write the story itself in one paragraph, not a command like “make me 7 frames”. The number of frames and the hero are chosen in “Settings”;
• Assembly. The Assemble video button becomes active when at least two frames are animated; next to it you can see how many there are now. The current storyboard is saved automatically and sits in “My storyboards”; a new one is created with the New storyboard button.

Voice and voice clone. For voicing frames there is “Default voiceover” (for all frames) and “Voice” in “Frame settings” (for one). You can pick ready-made voices, voices from the ElevenLabs library and your own clones. A voice clone is created both on the site (“Озвучка” → “Клон голоса”) and in the mobile apps: it needs the Pro plan or above and 18+ confirmation. Details are on the Voiceover and voices page.

413 October 2026: image series in chat and iPhone

• An image series in one message. In chat, write, for example, “make 7 images” and send the descriptions. The assistant will show a plan, the number of images and the price in ⚡, and you start it with the “Create N” button. The finished series is assembled into a grid, with “Download all” at the bottom. You can request up to ten images at a time;
• Editing by number. Write “make the third one brighter”: the new image appears in a new cell of the same grid, and the original stays in place;
• iPhone, builds 215–217 (TestFlight). A soul memory screen where entries can be edited, the Vidu Q4 model, a new Studio player (repeat, sound, AirPlay, thumbnail strip), pull-down refresh in “Photo”, “Video” and “Voiceover”, and the “Photo→video” chip.

512 October 2026: “Photo to Video”

In Studio, in the “Templates” section, a Photo to Video shelf has appeared. Upload 3–6 photos in order and get a vertical video: each photo comes to life with camera movement, and the frames are joined end to end. Niches: real estate, restaurant and cafe, cars, hotel and rentals, beauty salon, shop. You can choose the pace (“Calm” or “Dynamic”). Details are on the Studio templates page.

611 October 2026: documents on iPhone

• The iPhone app now has “My documents”: a cloud of files, with Open and Download on each card;
• “Download as…” was added, to save a chat reply in the format you need;
• Large files are now uploaded in parts, so a long upload doesn't break off halfway.

710 October 2026: “AI apps” in a new look

• The AI apps catalog has been redesigned: one large “Recommended” card and two medium ones, category chips with counts, dense app rows, and search with the “/” key;
• In the iPhone app's chat, answer ratings, a “Share” link, “Model comparison”, “Scheduled tasks” and projects from the site have appeared.

88–9 October 2026: new models and chat on the site

Chat on the site was rebuilt (9 October). The Gibrid assistant answers about the service and your account: it can show your balance and recent generations, search the web, draw an image and connect you with support. Added: projects in chat, answer ratings with a reason, “Share” by link, “Comparison” of 2–3 models, video from chat with a price card and a start button, “Scheduled tasks”, and a “Support” section.

New models in Studio. On 8 October these were connected: HappyHorse 1.1, Wan 3.0 Prime, PixVerse V6 (reference), Seedream 5.0 Flash, Qwen Image 2.1, Nano Banana 2.1, as well as Grok Imagine Video 1.5 Preview, Gemini Omni 1.1 Flash, Wan 2.6 Flash, Kling 3.0 Omni (transformation mode), Wan 2.7 Image and Pro. On 9 October, Qwen Image 3, Kling 2.1 (Master and Standard) and Wan 2.2 Turbo were added. The price of each model is shown in its card in ⚡.

926–29 September 2026: apps

• Mac, version 0.8.1: a new DMG build and a new “background” update: the app downloads the fresh version itself and offers a restart;
• iPhone, build 200 in TestFlight;
• Android: 0.8.1 (26 September), 0.8.2 (27 September), 0.8.3 (29 September). The APK file is at gibrid.ai/download/android.
• Where to download each app is on the Apps page.

Related