# Bad Times AI models and capabilities

## Prepared MV workflow
Model: bytedance/seedance-2-5, provider: Kie. Bad Times uses one input photo; pet and consenting-adult variants are prepared. Other templates accept their configured roles with one photo per selected role. Dual-role templates allow either or both characters; empty roles keep the original. Burning Bridges requires its lead and accepts up to eight optional supporting photos, assigned by first appearance, left to right when simultaneous. Not every support may be visible in its short source. Rumpelstiltskin’s visitor has no visible face; the workflow must not invent one. Motion and lip timing come from the selected reference; the source soundtrack is restored separately. Paid execution is disabled during acceptance. No free personalized preview is implemented. This model is not a facial-identity guarantee.

## Template pages
Server-owned template IDs pin reference video, role mapping, quote, aspect, duration and soundtrack. The same reserve → submit once → poll/callback → restore audio → store in private R2 → settle/refund lifecycle is used for every template. Mocked regression tests do not verify real paid output quality. Paid execution is disabled.

| Template | Photo roles | Seconds | Ratio | Reserve credits | Page |
| --- | --- | --- | --- | --- | --- |
| Bad Times · Pet | 1 | 10 | 9:16 | 300 | https://badtimes.app/ |
| Bad Times · People | 1 | 10 | 9:16 | 300 | https://badtimes.app/ |
| Bulin 47 | 1 | 10 | 9:16 | 300 | https://badtimes.app/templates/bulin-47 |
| Alors On Danse | 2 | 10 | 16:9 | 300 | https://badtimes.app/templates/alors-on-danse |
| Hotel Lobby | 2 | 10 | 16:9 | 300 | https://badtimes.app/templates/hotel-lobby |
| Rumpelstiltskin | 2 | 8 | 16:9 | 300 | https://badtimes.app/templates/rumpelstiltskin |
| Burning Bridges | 9 | 8 | 16:9 | 300 | https://badtimes.app/templates/burning-bridges |
| Zombie Couple | 2 | 10 | 16:9 | 300 | https://badtimes.app/templates/zombie-couple |
| Ozil Goal | 1 | 10 | 9:16 | 300 | https://badtimes.app/templates/ozil-goal |
| My Ride Is Here | 1 | 10 | 9:16 | 300 | https://badtimes.app/templates/my-ride-is-here |
| Smalltown Boy | 2 | 10 | 9:16 | 300 | https://badtimes.app/templates/smalltown-boy |

## Image and video tools
The inherited model catalog and quote engine are retained. Fal credentials are configured, but paid requests remain disabled. Browser image operations run locally; saving files to an account requires sign-in and enabled storage mutations.

## Input and usage limits
JPG, PNG or WebP, up to 10 MB each. MV backend requires 300–6000 px per side and aspect ratio 0.4–2.5. One subject per role. Do not upload public figures, children, non-consenting adults or deceptive impersonations. No guarantee of commercial rights to third-party music, reference footage or trademarks.

## Configured general video catalog (execution disabled)

Rates come from the same quote functions as the workspace. Defaults are examples; inspect the quote for your actual input durations, resolution and references.

### Genjutsu Motion Transfer (genjutsu-motion-transfer)
- Modes: reference.
- Inputs: reference: 8 images / 1 videos / 0 audio, 12 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 4–30 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio off.
- Prompt: optional, maximum 10000 characters.
- Quote: From 128 credits · default 4s input · 480p.


### Genjutsu Object Swap (genjutsu-object-swap)
- Modes: reference.
- Inputs: reference: 8 images / 1 videos / 0 audio, 12 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 4–30 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio off.
- Prompt: optional, maximum 10000 characters.
- Quote: From 128 credits · default 4s input · 480p.


### Genjutsu Restyle (genjutsu-restyle)
- Modes: reference.
- Inputs: reference: 5 images / 1 videos / 0 audio, 12 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 4–30 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio off.
- Prompt: optional, maximum 10000 characters.
- Quote: Conservative budget 255 credits · default 4s input · 480p · no character images; +100 credits/image.


### MiniMax H3 Max Turbo (minimax-h3-max-turbo)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480P, 768P, 1080P.
- Duration: 5–15 seconds.
- Defaults: text, 5s, 480P, 16:9, audio on.
- Prompt: required, maximum 50000 characters.
- Quote: From 7 credits · default 5s · 480P.


### MiniMax H3 (minimax-h3)
- Modes: reference, text, image.
- Inputs: reference: 9 images / 3 videos / 3 audio, 12 total; text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 2K.
- Duration: 5–15 seconds.
- Defaults: reference, 5s, 2K, adaptive, audio off.
- Prompt: required, maximum 7000 characters.
- Quote: From 130 credits · default 5s · 2K.


### H3 Max (minimax-h3-max)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480P, 768P.
- Duration: 5–15 seconds.
- Defaults: text, 5s, 768P, 16:9, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 50 credits · default 5s · 768P.


### Seedance 2.5 (seedance-2.5)
- Modes: reference, text, image.
- Inputs: reference: 30 images / 10 videos / 10 audio, 50 total; text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p.
- Duration: 4–30 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 84 credits · default 4s · 480p.


### Seedance 2.0 Fast (seedance-2.0-fast)
- Modes: reference, text, image.
- Inputs: reference: 9 images / 3 videos / 3 audio, 12 total; text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p.
- Duration: 4–15 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 44 credits · default 4s · 480p.


### Seedance 2.0 Pro (seedance-2.0-pro)
- Modes: reference, text, image.
- Inputs: reference: 9 images / 3 videos / 3 audio, 12 total; text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p, 1080p, 4k.
- Duration: 4–15 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 55 credits · default 4s · 480p.


### Seedance 2.0 Mini (seedance-2.0-mini)
- Modes: reference, text, image.
- Inputs: reference: 9 images / 3 videos / 3 audio, 12 total; text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p.
- Duration: 4–15 seconds.
- Defaults: reference, 4s, 480p, adaptive, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 29 credits · default 4s · 480p.


### Seedance 1.5 Pro (seedance-1.5-pro)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 4, 8, 12 seconds.
- Defaults: text, 4s, 480p, 16:9, audio off.
- Prompt: required, maximum 20000 characters.
- Quote: From 5 credits · default 4s · 480p.


### Wan 2.7 (wan-2.7)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 720p, 1080p.
- Duration: 2–15 seconds.
- Defaults: text, 5s, 720p, 16:9, audio off.
- Prompt: required, maximum 5000 characters.
- Quote: From 50 credits · default 5s · 720p.


### Wan 2.6 (wan-2.6)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p, 1080p.
- Duration: 5, 10, 15 seconds.
- Defaults: text, 5s, 720p, 16:9, audio off.
- Prompt: required, maximum 1500 characters.
- Quote: From 50 credits · default 5s · 720p.


### Grok Imagine (grok-imagine)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p.
- Duration: 6, 10 seconds.
- Defaults: text, 6s, 720p, 16:9, audio off.
- Prompt: required, maximum 5000 characters.
- Quote: From 42 credits · default 6s · 720p.


### Kling 3.0 (kling-3.0)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 720p.
- Duration: 3–15 seconds.
- Defaults: text, 3s, 720p, 16:9, audio off.
- Prompt: required, maximum 2500 characters.
- Quote: From 26 credits · default 3s · 720p.


### PixVerse Extend (pixverse-extend)
- Modes: reference.
- Inputs: reference: 0 images / 1 videos / 0 audio, 12 total.
- Resolutions: 360p, 540p, 720p, 1080p.
- Duration: 5 seconds.
- Defaults: reference, 5s, 720p, adaptive, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 20 credits · default 5s · 720p.


### ByteDance Video Upscaler (bytedance-video-upscaler)
- Modes: reference.
- Inputs: reference: 0 images / 1 videos / 0 audio, 12 total.
- Resolutions: 1080p, 2k, 4k.
- Duration: 5 seconds.
- Defaults: reference, 5s, 1080p, adaptive, audio off.
- Prompt: optional, maximum 2000 characters.
- Quote: From 4 credits · default 5s input · 1080p.


### Bria Video Background Removal (bria-video-background)
- Modes: reference.
- Inputs: reference: 0 images / 1 videos / 0 audio, 12 total.
- Resolutions: source.
- Duration: 5 seconds.
- Defaults: reference, 5s, source, adaptive, audio off.
- Prompt: optional, maximum 2000 characters.
- Quote: From 70 credits · default 5s input · source.


### PixVerse Person Swap (pixverse-swap)
- Modes: reference.
- Inputs: reference: 1 images / 1 videos / 0 audio, 12 total.
- Resolutions: 360p, 540p, 720p.
- Duration: 5 seconds.
- Defaults: reference, 5s, 720p, adaptive, audio off.
- Prompt: optional, maximum 2000 characters.
- Quote: From 20 credits · default 5s input · 720p.


### Sync Lip Sync 2 (sync-lipsync-2)
- Modes: reference.
- Inputs: reference: 0 images / 1 videos / 1 audio, 12 total.
- Resolutions: source.
- Duration: 5 seconds.
- Defaults: reference, 5s, source, adaptive, audio off.
- Prompt: optional, maximum 2000 characters.
- Quote: From 25 credits · default 5s input · source.


### Sync Talking Photo (sync-talking-photo)
- Modes: reference.
- Inputs: reference: 1 images / 0 videos / 1 audio, 12 total.
- Resolutions: source.
- Duration: 5 seconds.
- Defaults: reference, 5s, source, adaptive, audio off.
- Prompt: optional, maximum 2000 characters.
- Quote: From 67 credits · default 5s input · source.


### Hailuo 02 Pro (hailuo-02-pro)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 1080p.
- Duration: 6 seconds.
- Defaults: text, 6s, 1080p, 16:9, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 48 credits · default 6s · 1080p.


### Hailuo 02 Standard (hailuo-02-standard)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 768p.
- Duration: 6, 10 seconds.
- Defaults: text, 6s, 768p, 16:9, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 27 credits · default 6s · 768p.


### Hailuo 2.3 Pro (hailuo-2.3-pro)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 1080p.
- Duration: 6 seconds.
- Defaults: text, 6s, 1080p, 16:9, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 49 credits · default 6s · 1080p.


### Hailuo 2.3 Standard (hailuo-2.3-standard)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 768p.
- Duration: 6, 10 seconds.
- Defaults: text, 6s, 768p, 16:9, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 28 credits · default 6s · 768p.


### Seedance 1.0 Pro (seedance-1.0-pro)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 2–12 seconds.
- Defaults: text, 5s, 1080p, 16:9, audio off.
- Prompt: required, maximum 2000 characters.
- Quote: From 61 credits · default 5s · 1080p.


### Wan 2.5 (wan-2.5)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 1 audio, 1 total; image: 1 images / 0 videos / 1 audio, 2 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 5, 10 seconds.
- Defaults: text, 5s, 1080p, 16:9, audio on.
- Prompt: required, maximum 1500 characters.
- Quote: From 75 credits · default 5s · 1080p.


### Kling 2.6 Pro (kling-2.6-pro)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 1080p.
- Duration: 5, 10 seconds.
- Defaults: text, 5s, 1080p, 16:9, audio on.
- Prompt: required, maximum 2500 characters.
- Quote: From 70 credits · default 5s · 1080p.


### Veo 3.1 (veo-3.1)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p, 1080p, 4k.
- Duration: 4, 6, 8 seconds.
- Defaults: text, 8s, 720p, 16:9, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 320 credits · default 8s · 720p.


### HappyHorse 1.0 (happyhorse-1.0)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p, 1080p.
- Duration: 3–15 seconds.
- Defaults: text, 5s, 1080p, 16:9, audio off.
- Prompt: required, maximum 2500 characters.
- Quote: From 140 credits · default 5s · 1080p.


### HappyHorse 1.1 (happyhorse-1.1)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p, 1080p.
- Duration: 3–15 seconds.
- Defaults: text, 5s, 1080p, 16:9, audio off.
- Prompt: required, maximum 2500 characters.
- Quote: From 90 credits · default 5s · 1080p.


### Grok Imagine 1.5 (grok-imagine-1.5)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 480p, 720p, 1080p.
- Duration: 1–15 seconds.
- Defaults: text, 6s, 720p, 16:9, audio on.
- Prompt: required, maximum 4096 characters.
- Quote: From 84 credits · default 6s · 720p.


### H3 Max Styles (h3-max-styles)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 768P.
- Duration: 5–15 seconds.
- Defaults: text, 5s, 768P, 16:9, audio on.
- Prompt: required, maximum 49868 characters.
- Quote: From 40 credits · default 5s · 768P.


### H3 Max Camera Control (h3-max-camera-control)
- Modes: image.
- Inputs: image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 480P, 768P, 1080P.
- Duration: 3–15 seconds.
- Defaults: image, 5s, 768P, 16:9, audio off.
- Prompt: optional, maximum 50000 characters.
- Quote: From 40 credits · default 5s · 768P.


### H3 Max Lip Sync (h3-max-lip-sync)
- Modes: reference.
- Inputs: reference: 1 images / 0 videos / 1 audio, 12 total.
- Resolutions: 480P, 768P, 1080P, 2K.
- Duration: 5 seconds.
- Defaults: reference, 5s, 768P, adaptive, audio on.
- Prompt: optional, maximum 2000 characters.
- Quote: From 40 credits · default 5s input · 768P.


### Gemini Omni (gemini-omni)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 1 images / 0 videos / 0 audio, 1 total.
- Resolutions: 720p.
- Duration: 3–10 seconds.
- Defaults: text, 8s, 720p, 16:9, audio on.
- Prompt: required, maximum 20000 characters.
- Quote: From 100 credits · default 8s · 720p.


### H3 LoRA (h3-lora)
- Modes: text, image.
- Inputs: text: 0 images / 0 videos / 0 audio, 0 total; image: 2 images / 0 videos / 0 audio, 2 total.
- Resolutions: 480P, 768P, 2K, 4K.
- Duration: 5–15 seconds.
- Defaults: text, 5s, 768P, 16:9, audio on.
- Prompt: required, maximum 50000 characters.
- Quote: From 38 credits · default 5s · 768P.


## Configured general image catalog (execution disabled)

### Seedream 5.0 Pro (seedream-5-pro)
- Modes: layer-decomposition, text, image. Maximum input images: 10.
- Resolutions: 1K, 2K.
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3.
- Prompt: required, maximum 5000 characters.
- Outputs: 1–6.
- Quote: Generate: 1K 7 · 2K 14 credits/image; extra edit references quoted in the workspace. Layer reserve: Auto/1K/1.5K 72 · 2K 144 credits; settles by actual layers with excess refunded.


### GPT Image 2.5 Flare (gpt-image-2.5-flare)
- Modes: text, image. Maximum input images: 16.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: 1:1, auto, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 2:1, 1:2, 3:1, 1:3, 21:9, 9:21, 5:4, 4:5.
- Prompt: required, maximum 32000 characters.
- Outputs: 1–10.
- Quote: 1K square: low 4 · medium 4 · high 6 · xhigh 10 · max 22; High square: 1K 6 · 2K 11 · 4K 18 credits/image. Other ratios and custom sizes are quoted in the workspace; Auto uses the supported ceiling..


### GPT Image 2.5 Sunburst (gpt-image-2.5-sunburst)
- Modes: text, image. Maximum input images: 16.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: 1:1, auto, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 2:1, 1:2, 3:1, 1:3, 21:9, 9:21, 5:4, 4:5.
- Prompt: required, maximum 32000 characters.
- Outputs: 1–10.
- Quote: 1K square: low 4 · medium 4 · high 6 · xhigh 10 · max 22; High square: 1K 6 · 2K 11 · 4K 18 credits/image. Other ratios and custom sizes are quoted in the workspace; Auto uses the supported ceiling..


### GPT Image 2 (gpt-image-2)
- Modes: text, image. Maximum input images: 16.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: auto, 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 2:1, 1:2, 3:1, 1:3, 21:9, 9:21, 5:4, 4:5.
- Prompt: required, maximum 20000 characters.
- Outputs: 1–4.
- Quote: Square/auto: 1K 19 · 2K 25 · 4K 38; non-square: 1K 22 · 2K 27 · 4K 41 credits/image.


### Nano Banana 2 (nano-banana-2)
- Modes: text, image. Maximum input images: 14.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: auto, 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.
- Prompt: required, maximum 20000 characters.
- Outputs: 1–4.
- Quote: 1K 8 · 2K 12 · 4K 16 credits/image.


### Nano Banana Pro (nano-banana-pro)
- Modes: text, image. Maximum input images: 14.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9.
- Prompt: required, maximum 20000 characters.
- Outputs: 1–4.
- Quote: 1K 15 · 2K 15 · 4K 30 credits/image.


### Nano Banana (nano-banana)
- Modes: text, image. Maximum input images: 10.
- Resolutions: workspace default.
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9.
- Prompt: required, maximum 20000 characters.
- Outputs: 1–4.
- Quote: default 4 credits/image.


### Seedream 5 Lite (seedream-5-lite)
- Modes: text, image. Maximum input images: 10.
- Resolutions: 2K, 3K.
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9.
- Prompt: required, maximum 5000 characters.
- Outputs: 1–4.
- Quote: 2K 4 · 3K 4 credits/image.


### Seedream 4.5 (seedream-4-5)
- Modes: text, image. Maximum input images: 10.
- Resolutions: 2K, 4K.
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9.
- Prompt: required, maximum 3000 characters.
- Outputs: 1–4.
- Quote: 2K 4 · 4K 4 credits/image.


### Seedream V4 (seedream-v4)
- Modes: text, image. Maximum input images: 10.
- Resolutions: 1K, 2K, 4K.
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 5:4, 4:5, 21:9.
- Prompt: required, maximum 5000 characters.
- Outputs: 1–4.
- Quote: 1K 4 · 2K 4 · 4K 4 credits/image.


### Bria RMBG 2.0 (bria-background-remove)
- Modes: image. Maximum input images: 1.
- Resolutions: original.
- Aspect ratios: auto.
- Prompt: optional, maximum 0 characters.
- Outputs: 1.
- Quote: original 4 credits/image.


### Illusion Diffusion (illusion-diffusion)
- Modes: image. Maximum input images: 1.
- Resolutions: 1K.
- Aspect ratios: 1:1.
- Prompt: required, maximum 2000 characters.
- Outputs: 1.
- Quote: 1K 4 credits/image.


### Z-Image (z-image)
- Modes: text. Maximum input images: 0.
- Resolutions: workspace default.
- Aspect ratios: 1:1, 4:3, 3:4, 16:9, 9:16.
- Prompt: required, maximum 1000 characters.
- Outputs: 1.
- Quote: default 4 credits/image.


### Nano Banana 2 Lite (nano-banana-2-lite)
- Modes: text. Maximum input images: 0.
- Resolutions: 1K.
- Aspect ratios: auto, 1:1, 21:9, 16:9, 3:2, 4:3, 5:4, 4:5, 3:4, 2:3, 9:16, 4:1, 1:4, 8:1, 1:8.
- Prompt: required, maximum 50000 characters.
- Outputs: 1–4.
- Quote: 1K 5 credits/image.


### GPT Image 1.5 (gpt-image-1.5)
- Modes: text, image. Maximum input images: 10.
- Resolutions: 1K.
- Aspect ratios: 1:1, 3:2, 2:3.
- Prompt: required, maximum 32000 characters.
- Outputs: 1–4.
- Quote: 1K 14 credits/image.

Additional configured workflows below remain subject to the same disabled paid-execution boundary.

## Independent workflows

### Opus 5.5 · Motion Graphics (claude-opus-5-5-motion)

- Workspace: https://badtimes.app/opus-5-5-video-generator. Model: anthropic/claude-opus-5.5.
- Required: a motion brief (up to 6000 characters). Optional: up to 4 approved image elements, a confirmed website brief or a saved version to revise.
- Duration: 10, 15, 20, 30 seconds. Ratios: 16:9, 9:16, 1:1. Styles: sketch, motion.
- Defaults: 20s, 720p, 16:9, sketch. Silent motion graphics with supported English visible text, not photorealistic video; it produces a validated scene, with browser preview and MP4 export at 720p / 1080p.
- Billing unit: one create or revision request, not video seconds. The input-dependent quote reserves credits before submission (configured maximum 92); receipt-based settlement is capped at that reservation and returns the difference. Preview, saved-version selection and MP4 export do not create another LLM charge.

### Humanize text

- Workspace: https://badtimes.app/humanize-ai. Model: google/gemini-2.5-flash.
- Required: 1–12000 characters of source text. Tones: natural, friendly, professional; default natural. Output: rewritten text in the original language; no AI-detector guarantee.
- Billing unit: one rewrite. The same input-dependent quote function used by the workspace reserves 3 credits for the example input “Example text to rewrite.” with default tone. Longer or different inputs can change the quote. Receipt-based settlement is capped at the reservation and releases unused credits.

### Prompt Seen image-to-prompt

- Workspace: https://badtimes.app/prompt-seen. Model: google/gemini-2.5-flash.
- Required: one image/jpeg / image/png / image/webp image up to 10 MiB. The workspace normalizes it to JPEG with maximum 1024px per side before analysis. Optional guidance: up to 1000 characters. Styles: detailed, concise, cinematic; default detailed. Output: an English image-generation prompt describing visible content, not recovery of an original hidden prompt.
- Billing unit: one image-to-prompt request; reserve 2 credits and settle by the receipt, capped at the reservation, releasing unused credits.

### Seedream 5.0 Pro layer decomposition

- Workspace: https://badtimes.app/seedream-5-0-pro. Required: one image; input formats: image/jpeg, image/png, image/webp, image/bmp, image/tiff, image/gif; maximum 31457280 bytes. Layer-size options: auto, 1K, 1.5K, 2K. Output: separated image layers.
- Billing unit: returned layers. Reserve for 16 outputs: auto 72 credits; 1K 72 credits; 1.5K 72 credits; 2K 144 credits. Final settlement uses the same layer quote function and returned-layer count, capped at the reserved budget, with the difference returned.

These workflows require explicit submission and sufficient credits. Definitive failures release the reservation; requests awaiting reconciliation may remain reserved until their outcome is known. Credit returns are separate from cash refunds.