We’re creating the Figma for filmmakers. We’ve tackled all the challenges of browser-based video editing in order to bring AI-generated videos to life.
Very proud 🫶
MITO AI is here.
MITO AI is the tool we wish we had as filmmakers.
It is not a prompt playground or a template generator.
It is a control room for people who think in scenes, boards and timelines.
You bring a story, references and intent.
The tool helps you explore shots, workflows and versions faster than a traditional pipeline, without compromising taste.
• Cinematic boards in minutes.
• Keep your style consistent across images, scenes and videos.
Designed by a studio that actually uses it every day.
Join the early access list and help shape the tool: https://t.co/kDeMkHPKzC
Now 75% off all plans + unlimited Gemini Omni until July 7.
From idea to finished video in a few minutes.
Built with Gemini Omni Flash on MITO.
The discount link is in the first comment.
La IA va a contribuir bastante a la evolución de los estándares de HTML y CSS
Ya no hablo solo de crear aplicaciones o landings. Las tools de AI lo usan para crear presentaciones, ilustraciones, infografías, ....
New block in Notion: HTML.
Build interactive HTML right on your Notion page. Ask AI to turn your content into interactive explainers, prototypes, or diagrams.
Share with your team to use and tinker together.
Introducing PhoneFlow, the Open Source Voice Agent Platform 🚀
Lately Voice AI has been a very hot space:
In Spain alone there have been several major funding rounds recently, including HappyRobot ($44M), Prosper AI ($30M), Orbio ($21M), and Murphy AI ($15M).
If we take a broader look globally, there is an even more diverse set of players: Sierra (950M), Elevenlabs (500M), Parloa (350M), Decagon (250M), PolyAI (86M) Bland (50M), Vapi (50M), Synthflow (20M), Retell, etc
Although they have very different value propositions and GTM strategies (ranging from self-serve product led growth to FDE-driven enterprise sales motion or building for any use case to focusing on a single vertical) the core product behind most of them is actually quite similar: a platform to create voice agents by connecting nodes on a canvas with version control and deploy them to handle inbound and outbound phone calls.
Despite how crowded the space has become, I realized there wasn't an established open source alternative. So I decided to build one.
So far I had a lot of fun working on it. I got to solve some nerd sniping problems like:
- Transforming a mutable graph formed by different kinds of nodes connected through edges into a runtime state machine that dynamically updates the agent prompts depending on the current node and creates tools on the fly to transition through the outgoing edges when transition conditions are met
- A version control system with immutable snapshots of the agents to be able to iterate with confidence with full traceability of the changes
- A consumption based credit system that charges per call minute depending on the STT, LLM, TTS providers and models selected for each deployed agent version
- A basic templating engine to support dynamic variables in prompts
- Running and connecting the agents to handle real inbound and outbound phone calls
There are still a lot of features missing that I will continue to implement during my free time but I think this initial version is complete enough to put it out there.
Check out the GitHub repository, give it a star if you like it and join the waitlist to get notified when I release the platform generally available (link in comments).
Kling 3.0 in MITO AI gives you full creative control over every scene. Use our Infinite Canvas and Timeline to get the most out of these features:
> Sync Your Characters' Lips Perfectly: Use Lip Sync to match dialogue across multiple languages (English, Spanish, Japanese and more) without losing expressiveness or natural feel.
> Direct Real Physical Actions: Push beyond simple movements. Kling 3.0 handles complex physical interactions like collisions, weight, fabric and water, keeping every action grounded and believable.
> Master Complex Camera Movements: Choreograph dynamic shots with precision: dolly-ins, crane lifts, orbiting moves and seamless rack focuses, all generated with cinematic consistency.
> Layer Sound Effects on the Timeline: Bring your scenes to the Timeline and add SFX directly: footsteps, ambient textures, impacts, synced frame-by-frame to your visuals.
Try it now: https://t.co/EtMKDdeOZo
Es increíble la de capacidades que se van desbloqueando poco a poco conforme van saliendo nuevos modelos
La producción de video igual no está a la misma altura que la generación de código. Yo lo vivo bastante en el día a día, creo que de momento los developers tenemos mejores y más avanzadas herramientas que los creativos
La de locuras que se podrán hacer cuando los modelos de generación de imagen/video/audio estén a la misma altura que los modelos de texto que tenemos ahora
Achieving flawless character continuity across different shots is seamless with Seedance 2.0 in MITO AI. Leverage our Infinite Canvas and Timeline with these essential tips.
> Map Your Sequence on the Infinite Canvas: Visually organize up to 12 inputs side-by-side to lock in your character, lighting, and style before generating.
> Master the Narrative Multi-Shot: Direct a complete cinematic arc from Wide Shot to Close-Up and Closing Shot, keeping continuity perfectly consistent across every cut.
> Refine the Flow on the Timeline: Bring your generated shots straight to the Timeline to check the continuity in real time, trim transitions, and polish the final sequence.
Try it now: https://t.co/x080zrgm2i
Termino ahora de ver el WWDC porque levantar hierro siempre es más importante
Sorprendido para BIEN la verdad. El 75% de la conferencia centrada en AI
Apple no quiere entrar en la pelea en el barro por los modelos, lo que creo que es acertado.
Ha hecho lo que se da bien, conectar cosas (pero esta vez de verdad porque en la versión anterior 💩)
Termino ahora de ver el WWDC porque levantar hierro siempre es más importante
Sorprendido para BIEN la verdad. El 75% de la conferencia centrada en AI
Apple no quiere entrar en la pelea en el barro por los modelos, lo que creo que es acertado.
Ha hecho lo que se da bien, conectar cosas (pero esta vez de verdad porque en la versión anterior 💩)
Muchos casos de uso tienen overlap con ChatGPT o Claude pero para otros (mandar mensajes, modificar calendario, etc…) no va a tener rival por muchas integraciones y plugins que tengan los demás
La consistencia es un dolor cuando produces video con IA
En cada toma que se produce hay que lidiar con elno-determinismo de los LLMs y con el contexto para que todas las tomas tenga el mismo look
Si no luego al juntar todas las tomas la cosa parece un Frankenstein
En MITO estamos trabajando en esto y por ello presentamos Character Builder.
La idea es que puedas construir tus personajes y que tengas una plantilla que le puedas enchufar al modelo en cada generación para que mantenga la consistencia durante todo el film
Si haces filmación de video tradicional, puedes utilizar MITO para tu trabajo de preproducción y crear plantillas de los actores para luego utilizarlos para hacer algunas tomas con IA
The spot is below.
If you have a brief, a script, or just an idea, https://t.co/ZKnDukkpQm closes the gap between what you imagine and what you deliver.
Try it with the prompt in the first comment 👇
Alguna cosa interesante si ha salido.
PALABRAS POR ANUNCIO 🔠
Hay un crecimiento constante desde 2017 hasta 2025. No significa que los anuncios sean más largos, si no que se habla más en los anuncios.
Mi teoría es que pasamos cada vez más tiempo con el móvil mientras consumismo TV. Ya ni te digo si encima están en anuncios.
Todo lo que no hables en el anuncio, se pierde, ya que la gente no está mirando a la tele.
PD. He conseguido más anuncios del 2017