How Gabo Produces One-Hour AI Films With Higgsfield
2026-09-22 · other · Higgsfield, Gabo · original
Topic: Productions (case studies) (also: Video generation & shot direction, Assets, references & consistency, Failure modes & fixes, Studio work & business, Pre-production & story) · mode: Case study · tags: one creator's experience, guest or collaboration
This Higgsfield case study profiles Gabo of El Libro Sagrado, who makes feature-length AI films on history and myth, each one-hour film in 20 days to a month. On The Chronicles of Enoch he designed characters before the script, built references on a plain charcoal-grey background and baked weather and light into environment images, then moved from Seedance 2.0 to 2.5 and now works mainly in Cinema Studio 4.0. His prompts lead with the action, he keeps one model per film and checks each clip's final frame before the next, though many scenes still need repeated attempts and manual editing.
What is said
- Watch the Work, and the programme open to apply to
What is said
★ — how much it matters to the source.
- ★★★ Gabo is the creator behind El Libro Sagrado, a channel that turns historical and mythological stories into feature-length AI films.
- ★★ Views on some of Gabo's El Libro Sagrado films reach up to 6 million.
- ★★ The Chronicles of Enoch, the series Gabo made with Higgsfield, is available on a dedicated channel, Realframe Studios.
- ★★ Since joining Higgsfield's Creator Partnership Program (CPP), Gabo has scaled both the quality and the volume of his output significantly.
- ★★ Gabo's first experiments were short videos for TikTok, Facebook and Instagram on historical events, medieval stories, unusual facts and visually striking characters. · TikTok, Instagram
- ★★ Gabo first used Midjourney for images and Hailuo to animate them, later Krea AI, FLUX and ChatGPT; each project posed a new problem: making a still move, connecting two shots, keeping a character recognisable, sustaining a whole story. · Hailuo, FLUX, ChatGPT
- ★ Interstellar, Django Unchained, The Odyssey, Breaking Bad, Dark and the work of Quentin Tarantino shaped how Gabo builds tension, develops characters and connects several storylines.
- ★★ Gabo has produced around 50 AI films.
- ★★★ Gabo's production time for a one-hour AI film is 20 days to one month.
- ★★ The initial planning of one of Gabo's one-hour films (research, script, structure, characters) takes 4 to 5 days.
- ★★ Gabo reports a view increase of nearly five times on some releases of The Chronicles of Enoch.
- ★★ Gabo's higher-quality output brought more reach, more mentorship requests and more interest from people who wanted to learn from or work with him.
- ★★ Higgsfield says that in long-form AI filmmaking each shot is generated separately, so character appearance, lighting, backgrounds and body positions can change between clips, movement looks unnatural without direction, and environments drift across dozens of scenes. · character drift
- ★ Higgsfield says editing often has to hide generation errors in a long AI film, disconnected tools add friction at every handoff, and scenes needing several generations add significant time.
- ★★ Through the Creator Partnership Program Gabo gets monthly credits, access to emerging tools and early features, production resources, and a community where creators and programme leads exchange knowledge directly.
- ★★ With Creator Partnership Program access Gabo could test more ideas before committing to one, refine difficult scenes across multiple generations and choose the right combination of models before starting a full film.
- ★★ Gabo says Higgsfield marked a before and after in his career, and the quality and number of projects he can create have grown enormously.
- ★★★ Gabo tests several models before production and then keeps the selected model for the whole film, avoiding a switch midway.
- ★★★ Gabo uses ChatGPT, Claude and custom agents to organise and refine prompts, not to invent the story. · ChatGPT, Claude
- ★★ Gabo establishes a film's visual identity during early planning, before full scene generation begins.
- ★★ The Chronicles of Enoch draws on the mythology of the Book of Enoch: the Watchers, angels who descend to Earth, the emergence of giants, and characters like Ainmira, who becomes one of their first mothers.
- ★★ Gabo built The Chronicles of Enoch as a cinematic adaptation of the Book of Enoch with new conflicts, original characters and five storylines unfolding at once.
- ★★★ Gabo usually starts with the characters before a complete script exists, designing faces, clothing, proportions, personalities and reference sheets first, since seeing them shows how they speak, what they want and how they relate. · character sheet
- ★★ For a story the scale of The Chronicles of Enoch, Gabo defines where every major character needs to end up and works backward to build the journey, leaving room for new ideas mid-production.
- ★★★ For character identity Gabo builds detailed character references with Nano Banana, GPT Image 2 and Seedream 5.0 on a plain charcoal-grey background; removing distractions helps keep face, hair, clothing and body proportions across scenes. · Nano Banana, GPT Image 2, Seedream 5.0, character sheet
- ★★ Gabo develops character, environment, atmosphere and movement as separate layers instead of asking one model to generate them all at once.
- ★★★ For environments Gabo builds the intended weather, lighting, colour palette and detail (rain, wet ground, heavy clouds, smoke or direct sunlight) into the reference image from the start. · location reference sheet
- ★★★ For movement and direction The Chronicles of Enoch began on Seedance 2.0 and shifted increasingly to Seedance 2.5 for stronger prompt interpretation, higher visual quality and more control in the frame; Gabo's newer projects run mainly in Cinema Studio 4.0. · Seedance 2.0, Seedance 2.5, Cinema Studio 4.0
- ★★★ Gabo's video prompts put the principal action and dramatic intention first, followed by camera position, lens, movement, lighting, depth of field, colour and environment.
- ★★ Gabo composes with the rule of thirds, layered depth, off-centre framing and characters in profile or partly from behind, moving away from the centred symmetry that reads as generic AI output. · AI look
-
★★★ Gabo prefers to describe what a person physically does when feeling an emotion.
“I prefer to describe what a person physically does when they feel an emotion.”
-
★★ To show fear, Gabo's microactions are a tense jaw, interrupted breathing, a slight step backward and eyes searching beyond the frame.
- ★★ To show sadness, Gabo's microactions are rapid blinking to hold back tears, a lowered gaze, irregular breathing and wiping the nose with a sleeve.
- ★★ To show tension, Gabo's microactions are tightened fingers, restrained movements, a delayed response and a brief shift in posture.
- ★★★ Gabo often captures the final frame of a clip and analyses the character's position, gaze, background and composition before generating the next shot.
- ★★ In dialogue scenes Gabo cuts to the listening character's reaction before returning to the speaker, a natural edit that also hides small differences between separate generations.
- ★★ Higgsfield's case study advises using ChatGPT to analyse the final frame of a clip before generating the next shot. · ChatGPT
- ★★ When realism is needed, the case study advises prompting "photo of an actor" instead of "character sheet". · character sheet
- ★★ The case study advises applying the 60–30–10 colour rule to generated scenes.
- ★★ Higgsfield says it did not replace the storytelling, cinematography or editing behind Gabo's films but gave him the infrastructure to push all three further.
- ★★★ In Gabo's long-form work some scenes still need many generation attempts, exact poses remain hard to control, generation delays affect the schedule, and storytelling, continuity and editing still need substantial manual work.
- ★★ Gabo's workflow is centred on Higgsfield but not limited to it: ChatGPT and Claude both play a role alongside it. · ChatGPT, Claude
- ★ The Creator Partnership Program behind Gabo's workflow is open to direct applications from filmmakers.
Threads in this video
- first on record · Should you switch models during a production? — Gabo tests several models before production and then keeps the selected model for the whole film, avoiding a switch midway.
- restate · Which LLM should be used to write video prompts? — Gabo uses ChatGPT, Claude and custom agents to organise and refine prompts, not to invent the story.
- refine · How do you write the script for an AI film? — Gabo usually starts with the characters before a complete script exists, designing faces, clothing, proportions, personalities and reference sheets first, since seeing them shows how they speak, what they want and how they relate.
- restate · How do you keep a character the same across shots? — For character identity Gabo builds detailed character references with Nano Banana, GPT Image 2 and Seedream 5.0 on a plain charcoal-grey background; removing distractions helps keep face, hair, clothing and body proportions across scenes.
- refine · How do you keep a location, product or prop the same across shots? — For environments Gabo builds the intended weather, lighting, colour palette and detail (rain, wet ground, heavy clouds, smoke or direct sunlight) into the reference image from the start.
- shift · Which video model should be used for realistic people and film shots? — For movement and direction The Chronicles of Enoch began on Seedance 2.0 and shifted increasingly to Seedance 2.5 for stronger prompt interpretation, higher visual quality and more control in the frame; Gabo's newer projects run mainly in Cinema Studio 4.0.
- refine · How do you write a prompt that a video model follows in every detail? — Gabo's video prompts put the principal action and dramatic intention first, followed by camera position, lens, movement, lighting, depth of field, colour and environment.
- refine · How do you get a cinematic look in a generated image or shot? — Gabo composes with the rule of thirds, layered depth, off-centre framing and characters in profile or partly from behind, moving away from the centred symmetry that reads as generic AI output.
- refine · How do you avoid the AI (plastic) look? — When realism is needed, the case study advises prompting "photo of an actor" instead of "character sheet".
- restate · How do you direct a character's acting in a video prompt? — Gabo prefers to describe what a person physically does when feeling an emotion.
- refine · How do you direct a character's acting in a video prompt? — To show fear, Gabo's microactions are a tense jaw, interrupted breathing, a slight step backward and eyes searching beyond the frame.
- restate · How do you make a clip longer? — Gabo often captures the final frame of a clip and analyses the character's position, gaze, background and composition before generating the next shot.
- refine · How do you keep a consistent colour palette and visual style across generated images? — The case study advises applying the 60–30–10 colour rule to generated scenes.
Mentioned
Central: El Libro Sagrado · Gabo · microaction · The Chronicles of Enoch
Also: People: Quentin Tarantino
Projects: Creator Partnership Program
Works: Book of Enoch · Breaking Bad · Dark · Django Unchained · Interstellar
Organizations: Facebook · Instagram · Krea · O Livro Sagrado · Realframe Studios · TikTok
Concepts: 60–30–10 color rule · AI look · character drift · character sheet · location reference sheet · rule of thirds
Characters: Ainmira
Products: ChatGPT · Cinema Studio 4.0 · Claude · FLUX · GPT Image 2 · Hailuo · Midjourney · Nano Banana · Seedance 2.0 · Seedance 2.5 · Seedream 5.0
← Previous: Best AI Video Platforms for Agencies in 2026: Client Workspaces, Team Seats, and Commercial Rights · Next → Who will win the $1,000,000?