One-time packs from $4.99

Halloween AI Prompts

Copy tested Halloween AI prompts for costume photos, pet portraits, couples, short videos and 3D figures. Each recipe states its model, mode and credit cost.
Warm desk flat lay with printed prompt cards, a small pumpkin, a 3D figure and a camera lens

How to use a prompt recipe

Every recipe below has the same job: tell the model what to keep, what to change, where the subject is, how the scene is lit and what to avoid. The costume is only one line. The preservation and exclusion lines are what keep the face and body together.

Copy a recipe, paste it into the generator, then change one variable at a time. If you change the costume, the scene and the lighting in the same run, you will not know which change caused a failure.

Image prompts

Midnight Braids

A gothic portrait with violet braids, silver accessories and a moonlit garden. Strong for a full costume edit.

Prompt recipe

Midnight Braids

Costume, hair and scene in one recipe. Keep the face and pose.

Use the uploaded photo as the primary reference. Preserve the subject's recognizable facial identity, facial proportions, skin tone, approximate age and natural anatomy. Transform the subject into an original gothic Halloween fashion portrait. Change the hairstyle into long violet braids decorated with delicate silver hair accessories. Replace the clothing with an elegant black-and-purple velvet outfit featuring subtle lace and metallic details. Place the subject in a moonlit Victorian garden with soft fog, glowing lanterns and dark autumn foliage. Use photorealistic editorial photography, realistic fabric textures, natural skin details and cinematic soft lighting. Keep the same number of people as the uploaded photo. Avoid altered facial identity, extra limbs, duplicated fingers, distorted anatomy, text, logos and watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Cosmic Princess

A celestial gown with a star field and a moon crown. The deep-blue palette is the anchor.

Prompt recipe

Cosmic Princess

A glowing gown and an observatory scene, with the face kept intact.

Edit the uploaded photo into an original celestial Halloween princess portrait. Preserve the subject's recognizable face, natural skin texture, body proportions and pose. Replace the outfit with a midnight-blue fantasy gown featuring silver embroidery, softly glowing star patterns and realistic flowing fabric. Add a delicate silver moon crown and luminous accessories. Place the subject inside a moonlit observatory with a starry sky visible through tall windows. Use elegant fantasy fashion photography, realistic fabric details, cinematic rim lighting and a deep-blue, silver and violet color palette. Keep the person recognizable. Avoid altered facial identity, extra people, distorted anatomy, text and watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Halloween Robot

A wearable robot costume, not a robot head. Useful for a tech or sci-fi party look.

Prompt recipe

Halloween Robot

Armor panels, blue LEDs and a believable costume fit.

Using the uploaded photo as reference, transform the subject into a futuristic Halloween robot costume portrait. Preserve the subject's recognizable face, skin tone, head proportions and original pose. Create a realistic wearable robot costume featuring metallic-looking armor panels, subtle blue LED accents and articulated shoulder pieces. The outfit should look like a believable costume worn by a real person, not a completely new mechanical character. Use dramatic blue-and-orange cinematic studio lighting and realistic material reflections. Do not replace the subject's face with a robot head. Avoid extra limbs, distorted hands, duplicated body parts, logos and text.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Pumpkin Pet

A pet portrait with a soft pumpkin costume. Keep the eyes clear and the species unchanged.

Prompt recipe

Pumpkin Pet

A cozy autumn studio setup for one pet.

Use the uploaded pet photo as the primary reference. Preserve the animal's original breed, fur color, distinctive markings, eye color, facial appearance and natural anatomy. Dress the pet in a soft pumpkin-inspired Halloween costume with a small green leaf accessory. Keep the costume fitted naturally without covering the animal's eyes or changing its species. Place the pet in a cozy autumn photography studio with warm lighting and miniature pumpkins. Create a photorealistic professional pet portrait with realistic fur texture and natural shadows. Keep exactly one animal and the correct number of legs. Avoid extra animals, unnatural paws, duplicated limbs, text and watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Cowboy and Angel

A two-person recipe with distinct costumes and positions. State that there are exactly two people.

Prompt recipe

Cowboy and Angel

A coordinated couple portrait from one two-person photo.

Use the uploaded photo as the primary reference. The photo contains exactly two people. Preserve both people's recognizable facial identities, facial proportions, skin tones, ages, body proportions and pose. Dress the person on the left in an original western cowboy costume with a dark hat, suede vest and leather belt. Dress the person on the right in an original angel costume with a soft ivory layered outfit, feathered shoulder details and a simple halo. Place the couple on a foggy wooden porch at dusk with warm lantern light and carved pumpkins. Use photorealistic editorial photography, realistic fabric texture and cinematic soft lighting. Keep exactly two people and do not swap their faces. Avoid extra people, extra limbs, duplicated hands, text, logos and watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Video prompts

Slow head turn

A short motion beat for a finished costume portrait. A head turn and drifting fog are enough.

Prompt recipe

Costume portrait: slow head turn

Upload the finished image first, then paste this motion prompt.

Use the provided image as the exact visual reference and starting composition. Create a continuous short cinematic Halloween portrait animation. Keep the subject's recognizable face, costume, hairstyle and body proportions consistent with the reference. The subject slowly turns their head toward the camera with a subtle mysterious expression. A gentle breeze moves a few loose hair strands and the edge of the clothing. Soft fog drifts through the background. The camera performs a slow smooth push-in. Maintain realistic motion, stable anatomy and consistent lighting. Do not change the person's identity, replace the costume, add characters or introduce sudden scene cuts. No text or watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

Haunted corridor

A text-only atmosphere clip. No face, no identity risk, useful as a background.

Prompt recipe

Haunted corridor atmosphere

A liminal hallway with one flickering light and a slow forward move.

Create a short cinematic Halloween atmosphere clip. A long empty corridor with damp carpet, yellowed wallpaper and a single flickering ceiling light. Thin fog stays close to the floor. The camera moves forward slowly and smoothly at eye level with a gentle handheld feel. The light flickers once, then holds. Keep the geometry stable and the perspective consistent. No people, no creatures, no jump scare, no text, no logos and no watermarks.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

3D prompt

Printable Halloween figurine

A 3D prompt is different: describe one object, its material and the silhouette. Avoid scenes and multiple characters.

Prompt recipe

Halloween figurine

Upload a sketch or prop photo. The output is a textured GLB, not a print-ready mesh.

Create a single Halloween figurine from the uploaded reference: a stylized jack-o'-lantern creature with a rounded body, a carved smiling face, a short stem and two small feet. Keep one centered subject with a clear silhouette and no background. Use a matte ceramic material with subtle surface wear, warm orange body color and dark recessed carvings. Preserve the reference shape and proportions. Avoid thin unsupported parts, text, logos and extra objects.
Generate with this prompt

The prompt is pre-filled in the generator. Review it and the credit cost before you start.

What to change first

If a recipe fails, change the source image before you rewrite the prompt. A sharper face, a cleaner background or a more separated pose fixes more failures than another paragraph of instructions. If the source is good, change one prompt variable: costume, scene or lighting.

Open the tab that matches the recipe

Credits and output

Image recipes cost 80 credits per run. Video recipes cost 150 credits for Veo3 Fast or 300 credits for Veo3. A 3D run costs 150, 300 or 500 credits at 512, 1024 or 1536 resolution. The generator shows the cost before submission, and a failed tracked task returns its reserved credits. See pricing for current packs.

Copy one recipe and test it

Do not test five prompts at once. Pick the recipe closest to your source photo, run it once, and inspect the face, hands and background before changing anything.

Open the generator

Frequently asked questions

Can I use these prompts in Gemini or another image tool?
The prompts are written as plain instructions, so you can paste them into another editor that accepts image references. Results will differ because each model handles identity, anatomy and text differently. The credit costs and mode labels on this page apply to this hosted generator.
Why do the prompts repeat what to preserve?
Generative editors tend to change more than you asked. A short "preserve the face, pose and number of people" block keeps the model anchored. The preserve block is usually more important than adding more costume detail.
Do longer prompts produce better results?
Not automatically. Long prompts help when every sentence adds a visual instruction. They hurt when they contradict each other or describe two different scenes. Keep one subject, one costume, one location and one lighting setup per run.
Can I sell an image made with these prompts?
You are responsible for the rights to the source photo and for how you use the output. Do not upload a photo you do not have permission to edit, and do not present a generated costume as a real product. See the terms of service for the service rules.
Which prompt should I test first?
Start with a background or atmosphere edit. It changes less of the image and teaches you how the model handles your photo. Move to a full costume change after you see the first result.

TRELLIS.2 Image-to-3D Blog

Best Free Online 3D Viewers in 2026: GLB, STL, OBJ, FBX

Oct 8, 2026

Best Free Online 3D Viewers in 2026: GLB, STL, OBJ, FBX

Six free ways to open a 3D file without installing much, compared on formats, upload, sign-in and limits, including the TRELLIS.2 browser viewer.

How to Check a 3D Mesh Before Printing

Oct 4, 2026

How to Check a 3D Mesh Before Printing

Check a generated mesh in the browser before you convert it to STL and send it to a slicer.

How to Convert a GLB to STL

Oct 4, 2026

How to Convert a GLB to STL

Convert a GLB draft to STL in the browser, see which data you keep, and check the triangles before a slicer.

How to Open a 3MF File in the Browser

Oct 4, 2026

How to Open a 3MF File in the Browser

Open a 3MF package in the browser, orbit the mesh, and check what the file does and does not carry before a separate print step.

How to Open an FBX File in the Browser

Oct 4, 2026

How to Open an FBX File in the Browser

Drop an FBX file in the browser, orbit the mesh, and see which scene data the preview does and does not show.

How to Open a GLB File in the Browser

Oct 4, 2026

How to Open a GLB File in the Browser

Open a GLB file online, including the textured draft this site downloads from one image, and check the estimated sides.

How to Open a glTF File in the Browser

Oct 4, 2026

How to Open a glTF File in the Browser

Open a glTF file in the browser, and see why a lone JSON file fails when the buffer or textures were left behind.

How to Open an OBJ File in the Browser

Oct 4, 2026

How to Open an OBJ File in the Browser

Open an OBJ file in the browser, see the mesh without its sidecar materials, and know when to convert from a GLB.

How to Open a PLY File in the Browser

Oct 4, 2026

How to Open a PLY File in the Browser

Open a PLY mesh or scan in the browser, including vertex color, and see what a points-only file will not show.

How to Open an STL File in the Browser

Oct 4, 2026

How to Open an STL File in the Browser

View an STL file online, orbit the triangles, and see why the preview has no color and no print profile.

How to Turn a Photo into a 3D Model

Oct 4, 2026

How to Turn a Photo into a 3D Model

Turn one photo into a downloadable GLB draft, then inspect the estimated sides before you convert or print.

Open-Source 3D AI in 2026: A Verification-First Guide

Jan 27, 2026

Open-Source 3D AI in 2026: A Verification-First Guide

Compare hosted trials and open-source image-to-3D models without invented rankings, then check current TRELLIS.2 facts and a cost-per-accepted-asset test.

ComfyUI and TRELLIS.2: A Safe Local Setup Checklist (2026)

Jan 27, 2026

ComfyUI and TRELLIS.2: A Safe Local Setup Checklist (2026)

Understand the official TRELLIS.2 requirements, third-party ComfyUI wrapper boundary, installation safety, model verification, workflow structure, and troubleshooting.

Open-Source Image-to-3D: TRELLIS.2 vs Hunyuan3D, TripoSR

Jan 27, 2026

Open-Source Image-to-3D: TRELLIS.2 vs Hunyuan3D, TripoSR

A source-based review of the open-weight image-to-3D releases you can run yourself: TRELLIS.2, Hunyuan3D 2.1, TripoSR and Stable Fast 3D, with their licences and hardware.

TRELLIS.2 to Blender: Import GLB, Inspect, Clean, and Export

Jan 27, 2026

TRELLIS.2 to Blender: Import GLB, Inspect, Clean, and Export

A practical TRELLIS.2-to-Blender workflow for importing the hosted GLB, checking transforms, geometry and PBR materials, cleaning topology, and exporting safely.

TRELLIS.2 Tutorial: From One Image to a Reviewed GLB

Jan 27, 2026

TRELLIS.2 Tutorial: From One Image to a Reviewed GLB

A step-by-step TRELLIS.2 guide covering input selection, 512/1024/1536 credit cost, tracked generation, GLB inspection, failure refunds, and destination-tool validation.

Can TRELLIS.2 Models Be 3D Printed? Mesh Validation

Jan 27, 2026

Can TRELLIS.2 Models Be 3D Printed? Mesh Validation

A practical, non-fabricated TRELLIS.2 print-readiness workflow covering manifoldness, holes, normals, wall thickness, scale, STL conversion, slicing, and test prints.

TRELLIS.2 in a VFX Pipeline: An Evaluation Checklist

Jan 27, 2026

TRELLIS.2 in a VFX Pipeline: An Evaluation Checklist

A transparent workflow for evaluating TRELLIS.2 output for VFX, previs, set dressing, and production handoff, without any fabricated artist testimonials.

Trellis 2 Quickstart: Your First Image-to-3D Draft

Jan 27, 2026

Trellis 2 Quickstart: Your First Image-to-3D Draft

A truthful five-minute TRELLIS.2 quickstart: sign in, buy credits, start a 512 job, follow live job status, inspect the result, and download a GLB draft.

TRELLIS.2 (trellis-2) Guide: Image-to-3D, PBR, GLB

Dec 18, 2025

TRELLIS.2 (trellis-2) Guide: Image-to-3D, PBR, GLB

A practical trellis-2 / trellis 2 / trellis.2 guide: what TRELLIS.2 is, how image-to-3D works, best inputs, key settings, and how to export/import GLB assets.

← Swipe →