Selendis Introduces Archon, a Video Foundation Model Built With Steve Aoki to Edit Raw Footage in Each Creator's Style
VP Land
Nov 9, 2026 · 2 min read
Selendis introduced Archon, a video foundation model built by Steve Aoki and industry experts that the company says turns raw footage into a finished edit in minutes, in a launch post on X. Archon powers personal AI video editors that Selendis says learn each creator's editing taste and get more accurate with every video edited.
Raw footage to a finished edit. Selendis claims Archon can "zero-shot" raw footage into a finished edit in minutes.
Taste comes from three inputs. The personal editors learn from decisions made in Selendis' multiplayer editor, a creator's past video catalogue and reference videos the creator cites.
Accuracy is meant to build per creator or team. According to Selendis, every video edited makes the editor more accurate for that creator or team.
Access starts with a waitlist. Selendis is in early access, and the launch post adds three free months for people who comment to request access.
Archon learns from a creator's catalogue, cited references and decisions made inside Selendis' editor
On its official site, Selendis describes itself as an AI lab building Archon to power personal AI video editors. The company says Archon adapts to individual creator taste, learns each user's editing style and can zero-shot video content in minutes.
Those editors are delivered through what Selendis calls an "AI-native multiplayer video editor." Selendis lists a creator's past video catalogue and the reference videos they cite alongside decisions made in that multiplayer editor as the signals the editors learn from.
The company frames that learning around individual creators and teams. Per Selendis, each edited video makes the editor more accurate for that creator or team, though it has not said how many edits that takes or what changes along the way.
Aoki is on the build team, but Selendis has not shown what sits under the "foundation model" label
Selendis credits Steve Aoki and unnamed industry experts with building Archon. A Selendis video on X shows Aoki alongside Gavin Wong, with both labeled as co-founders.
Steve Aoki and Gavin Wong, labeled as co-founders, in a Selendis video on X. Image: X (formerly Twitter)
The launch post and site describe what Archon is supposed to do but share no model architecture, training data, supported formats, output specs or pricing. One reply on the launch post, from Noah Morris, questioned whether Selendis "actually trained up any novel model from scratch."
Until Selendis publishes technical details, the "zero-shot" and "in minutes" descriptions remain company claims.
Early access reserves a username, with three free months for commenters
Selendis says it is in early access, and joining the Selendis waitlist reserves a username and a place in line. In a follow-up post, the company urged people to reserve their handle before someone else takes it.
The launch post also offers early access plus three months free to anyone who comments "SELENDIS" on it.
Pika released H3 Flex, an open-weight video model post-trained on MiniMax H3 that generates a 15-second clip with soundtrack in as little as 7 seconds, according to Pika's launch post. Pika built it for professional i…
Google DeepMind introduced Genie 4, a new world model in research preview that you can steer with text prompts while a simulation runs and that generates its own audio, from ambient sound to dialogue. Google's trusted…
Utopai Studios has launched PAI, a production intelligence platform for film and television that breaks a screenplay into scenes and shots, alongside Utopai X, a new video generation model that creates footage inside…