Experiments

Older models hallucinate images. GPT-4o writes them.

Old diffusion models like DALL·E 2 or Midjourney generate images by starting with random noise (a seed) and refining the whole canvas at once—much like a photo gradually developing in a darkroom. The final image “emerges” from the fog, loosely guided by the prompt.

OpenAI’s new models work

Storytelling, Sorrytelling, sorrytellers

All my art is based on a fantasy.

On commitment, on effort.

The cathedral, sign of unification that disregard the practical realities

On broken dreams.

On the realisation of a world.

Unifying with a manifesto

Cloches - Monaster - Ruinę / Appartement.

Route - sinueuse

Mer - Juste illustrative du temps agité.

Bateau - La fumé du bateau.

Violence - humaine. Ne montre les machines.

Resolution (Cloches)

It's all a strange paradigm

We are artists

And we live in a studio

We breathe, think, talk, work, love,

And we make,

Our mind has taught to our hand

What our heart had told to our eyes

— Siméon Artamonov

20240830-130517

It was a dream,

Erased from my memory

A quiet dream

So real

yet so elusive,

A forgotten moment,

Which wasn't

Это был сон,

Стертый из моей памяти,

Тихий сон,

Настолько реальный,

Но ускользающий,

Забытый момент,

Которого не было.

Demain - short generative AI clip

The boy who yearned for the mane event.

Experimentation in Gen AI short film making - Siméon Artamonov

Midjourney 6.0 / Luma AI - Dream Machine / Runway ML Gen-3 Alpha / Editing: Runway Project.

Voice: Eleven Labs / Music Gen: Sumo ai

Tomorrow, elsewhere,

In the city's atelier,

You will be there.

At the speed of dreams,

To