Krea AI — Image, Video & Enhancement Generation

Generate images, videos, upscale/enhance images, and train LoRA styles using the Krea.ai API. Supports 20+ image models (Flux, Imagen, GPT Image, Ideogram, Seedream...), 7 video models (Kling, Veo, Hailuo, Wan), and 3 upscalers (Topaz up to 22K).

IMPORTANT: Do NOT invent model names. Before generating, run

list_models.py

list_models.py --json

--model /generate/image/google/imagen-4-ultra

Usage

Run scripts using absolute path (do NOT cd to skill directory first):

Generate image:

uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "your description" --filename "output.png" [--model flux] [--width 1024] [--height 1024] [--api-key KEY]

Generate video:

uv run ~/.codex/skills/krea/scripts/generate_video.py --prompt "your description" --filename "output.mp4" [--model kling-2.5] [--duration 5] [--aspect-ratio 16:9] [--api-key KEY]

Enhance/upscale image:

uv run ~/.codex/skills/krea/scripts/enhance_image.py --image-url "https://..." --filename "upscaled.png" --width 4096 --height 4096 [--enhancer topaz] [--api-key KEY]

List available models:

uv run ~/.codex/skills/krea/scripts/list_models.py [--type image|video|enhance]

Run a multi-step pipeline:

uv run ~/.codex/skills/krea/scripts/pipeline.py --pipeline pipeline.json [--api-key KEY]

Check job status:

uv run ~/.codex/skills/krea/scripts/get_job.py --job-id "uuid" [--api-key KEY]

Important: Always run from the user's current working directory so files are saved where the user is working.

Default Workflow (draft → iterate → final)

Goal: fast iteration without burning CU on expensive models until the prompt is right.

• Draft (cheap/fast): use

  --model z-image

  or

  --model flux

  (3-5 CU, ~5s)

  uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "<draft prompt>" --filename "yyyy-mm-dd-hh-mm-ss-draft.png" --model flux
  

• Iterate: adjust prompt, keep trying with cheap models

• Final (high quality): switch to

  --model gpt-image

  or

  --model nano-banana-pro

  uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "<final prompt>" --filename "yyyy-mm-dd-hh-mm-ss-final.png" --model nano-banana-pro
  

Image Models (sorted by cost)

z-image

flux

flux-kontext

qwen

imagen-4-fast

ideogram-2-turbo

seedream-4

seedream-5-lite

flux-pro

nano-banana

imagen-3

imagen-4

runway-gen4

flux-pro-ultra

imagen-4-ultra

nano-banana-flash

ideogram-3

nano-banana-pro

gpt-image

Map user requests:

• "fast", "quick", "cheap" →

  flux

  or

  z-image

• "high quality", "best" →

  nano-banana-pro

  or

  gpt-image

• "text in image", "typography" →

  ideogram-3

• "photorealistic" →

  seedream-4

  or

  nano-banana-pro

• No preference →

  flux

Video Models

kling-1.0

kling-1.5

kling-2.5

veo-3

veo-3.1

hailuo-2.3

wan-2.5

Map user requests:

• "fast" →

  hailuo-2.3

• "cinematic", "high quality" →

  veo-3.1

• "with sound", "with audio" →

  veo-3

  with

  --generate-audio

• No preference →

  kling-2.5

Enhancers

topaz

topaz-generative

topaz-bloom

Image Generation Parameters

--model

flux

--prompt

--filename

--width

--height

--aspect-ratio

--resolution

--seed

--image-url

--style-id

--style-strength

--batch-size

--steps

--guidance-scale

--quality

--api-key

Video Generation Parameters

--model

kling-2.5

--prompt

--filename

--duration

--aspect-ratio

--start-image

--end-image

--resolution

--mode

--generate-audio

--api-key

Enhancement Parameters

--enhancer

topaz

--image-url

--filename

--width

--height

--enhancer-model

--creativity

--face-enhancement

--sharpen

--denoise

--api-key

API Key

Scripts check for API key in this order:

1. --api-key

   argument (use if user provided key in chat)

2. KREA_API_TOKEN

   environment variable

If neither is available, the script exits with an error message.

Preflight + Common Failures

Preflight:

• command -v uv

  (must exist)

• test -n "$KREA_API_TOKEN"

  (or pass

  --api-key

  )

Common failures:

• Error: No API key

  → set

  KREA_API_TOKEN

  or pass

  --api-key

• 402 Insufficient credits

  → top up compute units at https://krea.ai/settings/billing

• 402 This model requires a higher plan

  → model needs a paid plan upgrade at https://krea.ai/settings/billing

• 429 Too many requests

  → concurrent job limit reached; scripts auto-retry up to 3 times with backoff

• Job failed

  → check prompt for content moderation issues, try different wording

Filename Generation

Generate filenames with the pattern:

yyyy-mm-dd-hh-mm-ss-name.ext

• Timestamp: current date/time in

  yyyy-mm-dd-hh-mm-ss

  (24h format)
• Name: descriptive lowercase text with hyphens (1-5 words)
• Extension:

  .png

  for images,

  .mp4

  for videos

Examples:

• Prompt "A cyberpunk cat" →

  2025-11-23-14-23-05-cyberpunk-cat.png

• Prompt "waves on a beach" →

  2025-11-23-15-30-12-beach-waves.mp4

Prompt Handling

For generation: Pass user's description as-is to

--prompt

For image-to-image: Use

--image-url

--prompt

For video from image: Use

--start-image

--prompt

Preserve user's creative intent in all cases.

Output

• Scripts download the result and save it to the current directory
• Script outputs the full path to the generated file
• Do not read the image/video back — just inform the user of the saved path
• If

  --batch-size

  > 1, files are saved as

  name-1.png

  ,

  name-2.png

  , etc.

Examples

Quick draft image:

uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "A serene Japanese garden with cherry blossoms" --filename "2025-11-23-14-23-05-japanese-garden.png" --model flux

High quality final:

uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "A serene Japanese garden with cherry blossoms, golden hour lighting" --filename "2025-11-23-14-25-30-japanese-garden-final.png" --model nano-banana-pro --resolution 4K

Image-to-image edit:

uv run ~/.codex/skills/krea/scripts/generate_image.py --prompt "transform to watercolor painting style" --filename "2025-11-23-14-30-00-watercolor.png" --image-url "https://example.com/photo.jpg" --model nano-banana-pro

Generate video:

uv run ~/.codex/skills/krea/scripts/generate_video.py --prompt "A majestic eagle soaring over snow-capped mountains at sunrise" --filename "2025-11-23-15-00-00-eagle-mountains.mp4" --model veo-3 --duration 8 --generate-audio

Upscale image to 4K:

uv run ~/.codex/skills/krea/scripts/enhance_image.py --image-url "https://example.com/photo.jpg" --filename "2025-11-23-15-10-00-upscaled.png" --width 4096 --height 4096 --enhancer topaz

List models:

uv run ~/.codex/skills/krea/scripts/list_models.py --type image

Pipelines (Multi-Step Workflows)

Use

pipeline.py

Pipeline JSON Format

{
  "steps": [
    {
      "action": "generate_image | generate_video | enhance | fan_out",
      "model": "model-id",
      "prompt": "...",
      "filename": "base-name",
      "use_previous": true,
      "...": "any other model-specific params"
    }
  ]
}

Actions:

• generate_image

  — generate an image

• generate_video

  — generate a video

• enhance

  — upscale/enhance an image

• fan_out

  — run a sub-step for EACH result from the previous step (branching)

Special fields:

• use_previous: true

  — use the output URL(s) from the previous step as input

• fan_out

  has a

  step

  field containing the template to run per source URL
• In fan_out prompts,

  {i}

  is replaced with the iteration number (1, 2, 3...)

Example: Generate → 4 Angles → 4 Videos

Generate a concept image, create 4 angle variations, then animate each one:

{
  "steps": [
    {
      "action": "generate_image",
      "model": "flux",
      "prompt": "a red sports car on an empty highway, golden hour, cinematic",
      "filename": "car-concept"
    },
    {
      "action": "fan_out",
      "use_previous": true,
      "step": {
        "action": "generate_image",
        "model": "gpt-image",
        "prompt": "same red sports car, angle {i} of 4: front three-quarter view at angle {i}, professional automotive photography, studio lighting, white background",
        "filename": "car-angle-{i}"
      }
    },
    {
      "action": "fan_out",
      "use_previous": true,
      "step": {
        "action": "generate_video",
        "model": "kling-2.5",
        "prompt": "the red sports car slowly rotates on a turntable, smooth motion, studio lighting",
        "duration": 5,
        "filename": "car-spin-{i}"
      }
    }
  ]
}

Run it:

uv run ~/.codex/skills/krea/scripts/pipeline.py --pipeline car-pipeline.json

Example: Generate → Upscale → Animate with Audio

{
  "steps": [
    {
      "action": "generate_image",
      "model": "nano-banana-pro",
      "prompt": "a majestic dragon perched on a cliff overlooking a stormy ocean",
      "filename": "dragon"
    },
    {
      "action": "enhance",
      "use_previous": true,
      "enhancer": "topaz-generative",
      "width": 4096,
      "height": 4096,
      "creativity": 3,
      "filename": "dragon-4k"
    },
    {
      "action": "generate_video",
      "use_previous": true,
      "model": "veo-3",
      "prompt": "the dragon spreads its wings and roars, lightning strikes, waves crash below, epic cinematic",
      "duration": 8,
      "generateAudio": true,
      "filename": "dragon-epic"
    }
  ]
}

Example: Product Photography Pipeline

Generate hero shot → 4 style variations → upscale all:

{
  "steps": [
    {
      "action": "generate_image",
      "model": "gpt-image",
      "prompt": "minimalist perfume bottle, frosted glass, on marble surface, soft studio lighting, product photography",
      "quality": "high",
      "filename": "perfume-hero"
    },
    {
      "action": "fan_out",
      "use_previous": true,
      "step": {
        "action": "generate_image",
        "model": "gpt-image",
        "prompt": "same perfume bottle, variation {i}: 1=morning light with flowers, 2=dark moody with smoke, 3=underwater with bubbles, 4=floating in clouds",
        "filename": "perfume-mood-{i}"
      }
    },
    {
      "action": "fan_out",
      "use_previous": true,
      "step": {
        "action": "enhance",
        "enhancer": "topaz",
        "width": 4096,
        "height": 4096,
        "filename": "perfume-final-{i}"
      }
    }
  ]
}

Example: Inline Pipeline (no JSON file needed)

For quick pipelines, pass JSON directly:

uv run ~/.codex/skills/krea/scripts/pipeline.py --pipeline '{"steps":[{"action":"generate_image","model":"flux","prompt":"a cat astronaut","filename":"cat"},{"action":"enhance","use_previous":true,"enhancer":"topaz","width":4096,"height":4096,"filename":"cat-4k"}]}'

Building Pipelines for Users

When a user asks for something complex like "generate a product shot from 4 angles and make videos of each":

1. Write a pipeline JSON with the right steps
2. Save it to a

   .json

   file in the current directory
3. Run it with

   pipeline.py --pipeline file.json

4. Show the user the saved file paths when done

Tips:

• Use

  fan_out

  to branch — it runs the sub-step once per result from the previous step

• use_previous: true

  chains steps automatically
• Start cheap (

  flux

  ) for concept, switch to quality (

  gpt-image

  ,

  nano-banana-pro

  ) for finals
• Upscale as the last step before video — cheaper and better quality
• Use

  {i}

  in prompts inside

  fan_out

  to vary the prompt per iteration

Pipeline Advanced Features

Template variables — Use

{{variable}}

uv run ~/.codex/skills/krea/scripts/pipeline.py --pipeline template.json --var subject="red sports car" --var style="cinematic"

Pipeline JSON can then use

{{subject}}

{{style}}

Parallel fan_out — Add

"parallel": true

{
  "action": "fan_out",
  "use_previous": true,
  "parallel": true,
  "step": { "action": "enhance", "enhancer": "topaz", "width": 4096, "height": 4096, "filename": "upscaled-{i}" }
}

Control concurrency with

--max-parallel N

Resume interrupted pipelines — Use

--resume

.pipeline-state.json

use_previous

Dry-run — Use

--dry-run

Notifications — Use

--notify