Release questions: answered from the notes
Each question, when clips reached thirty seconds or which models took voice references, is answered with the dated entries that bear on it. As of 2026-09-25.
Answers are lists of entries in date order, each linked to the note.
An answer covers the notes read; a feature shipped without a note is not in it.
1In this section
- Can AI video models edit part of a clip? — Several notes describe edits to footage that exists rather than a new generation: Wan 2.1 VACE's local editing in May 2025…
- How long do AI video models stay in preview? — Where the notes give both dates, weeks to months.
- How many reference images do AI video models take? — The notes that give a number: Veo 3.1 up to three images, Vidu one to seven, Wan 2.7 up to five images or videos mixed, HappyHorse 1.0 up to nine…
- When did AI video clips reach 30 seconds? — Per generation, the notes read go from 10 seconds with Hailuo 02 in June 2025 to 15 with Wan 2.6 in December, 20 with Sora in March 2026, and 30 with Wan…
- When did AI video models add vertical video? — Runway opened vertical generation to all users on Gen-3 Alpha Turbo in September 2024 and added Expand Video for new aspect ratios that November.
- When did AI video models get 4K output? — Runway offered 4K upscaling inside Gen-3 Alpha in January 2025.
- When did AI video models get camera controls? — The first dated camera controls read are Firefly's angle, motion and zoom controls in its October 2024 beta and Runway's Camera Control with direction…
- When did AI video models open to developers? — The notes read date these openings: Runway's API in September 2024, Kling's API taking models by November, Nova Reel in Amazon Bedrock in December, Veo 2…
- When did AI video models reach 1080p? — The first dated 1080p note read is the Firefly Video Model's in February 2025, then Hailuo 02 in June and Veo 3's upscaling on Vertex AI in July.
- When did lip-sync arrive in AI video models? — Kling added lip-sync in December 2024, stretched it to 60 seconds in June 2025 and to several faces in September.
- Where else can you run Wan 3.0? — Alibaba Cloud released Wan 3.0 on 6 August 2026 and its faster Prime version on 20 August.
- Which AI video models add HDR or higher frame rates? — Luma's Ray3 generated native HDR in EXR files from September 2025.
- Which AI video models can extend a clip? — Extension appears early: Gen-3 Alpha to 40 seconds in total in September 2024, Gen-3 Alpha Turbo to 34, and Firefly's Generative Extend in October.
- Which AI video models copy a performance from a video? — Runway's Act-One drove a character from one performance video in October 2024 and Act-Two tracked head, face, body and hands in July 2025.
- Which AI video models generate sound with the picture? — In the notes read: Veo 3 from its preview in July 2025, the Wan 2.5 preview in September 2025 and later Wan versions, Vidu's joint audio and video output…
- Which AI video models let you set how much moves? — Among the notes read: Kling's Motion Brush on its 1.0 model in November 2024 marks where things move, Runway's Camera Control in November 2024 sets…
- Which AI video models let you set or clone a voice? — Kling 2.6 took a chosen voice in December 2025 and let elements carry a voice from March 2026.
- Which AI video models plan several shots in one clip? — Amazon's Nova Reel 1.1 made multi-shot videos from April 2025, joining 6-second shots.
- Which AI video models take a first and a last frame? — Among the notes read: Runway's Gen-3 Alpha from August 2024, Firefly in February 2025, PixVerse and Wan 2.1 in spring, Veo 2 on Vertex AI in June…
- Which AI video models take a reference video? — Several notes take video, not only pictures, as a reference: Firefly for composition from July 2025 and for camera motion from December, Wan 2.2 Animate…
- Which AI video release notes claim better motion? — Motion claims come with new models and updates: Runway's Gen-3 Alpha in July 2024, Wan 2.1 in January 2025 and Wan 2.2 in July, the Firefly Video Model's…
- Which outside models reached both Runway and Higgsfield? — Gemini Omni Flash, Wan 3.0 and four Seedance steps: 2.0 in 4K, 2.0 Mini, 2.5 and 2.5 at 1080p.