Higgsfield Prompting Cheat Sheet

Pick the preset and name the move in the prompt. Doing only one gets you approximately what you asked for.

TL;DR — THE SHORT VERSION

It is not a prompt-only tool. The camera move comes from a preset you select in the app.
Name the same move in the prompt text as well. That pairing is what makes results repeatable.
50+ named motion presets — crash zoom, bullet time, dolly, crane, FPV drone, Snorricam.
Your prompt's job is narrower: subject, identity, environment, action.
Do not describe a different camera move from the preset you selected.

SECONDARY SOURCING

This vendor publishes no consolidated prompting guide. Built from platform documentation and community testing — weaker than the vendor-documented sheets.

The formula

Preset (selected in app) + named move in text + subject + identity anchor + environment + action

Template:

[named move, matching your selected preset], [subject with identity anchors], [environment and time], [action across the clip], [look].

Worked example:

Slow dolly push-in, a ceramicist in her fifties with a flour-dusted apron, north-facing studio at dusk, centring clay and drawing the form upward, muted earth palette and window light only.

Breakdown — what each part does

Preset

Chosen in the interface. This is where the camera movement actually comes from.

Named move in text

Repeat the move by name in the prompt so text and preset agree.

Subject and identity

Reference upload plus explicit identity anchors — the model's strength is putting a specific person in the shot.

Environment and action

What surrounds the subject and what happens across the clip.

Does it take a negative prompt?

No dedicated negative field.NO

Handle exclusions the way you would on any prompt-only video model: describe the outcome you want. Because the camera is preset-controlled, most camera-related exclusions are unnecessary — you are not fighting the model for the move, you selected it.

What not to do

Do not describe an elaborate camera move in text while a different preset is selected — the two conflict and the result is neither.✕

Do not treat it as prompt-only; skipping the preset leaves its main feature unused.✕

Do not over-specify the camera in text beyond naming the selected move.✕

Do not omit identity anchors when using a face reference.✕

SOURCING NOTE

Higgsfield publishes a preset library rather than a consolidated prompting guide. The workflow described here is drawn from that library and platform documentation, so treat it as well-established practice rather than a vendor style manual.

The checklist

Before you send it:

1 — Preset selected in the app?□

2 — Same move named in the prompt text?□

3 — Text and preset describing the same move, not two different ones?□

4 — Identity anchored if using a reference?□

5 — Action described across the clip, not just a pose?□

SOURCE

Higgsfield camera-controls preset library, higgsfield.ai/camera-controls, and platform guides. Secondary sourcing — no consolidated vendor prompting guide is published. Checked 25 Aug 2026.

Model versions and vendor documentation both move. Re-read the source before relying on a specific number. Errors are logged at corrections.

Every other sheet in this set: all model cheat sheets.