Eden Gen

References and elements

Animate an approved still, keep a face consistent, match a product. Upload once, reuse everywhere: app, API and agents.

Three ways in

HowUse when
POST /v1/assetsYou have the bytes: multipart field file, or a raw body with Content-Type and X-Filename.
POST /v1/uploadsLarge files or agents: a one-time URL you PUT to with curl, valid 30 minutes, no API key in the command.
{ url } or { take_id }A public https link, or one of your finished takes. Private network addresses are refused.

The file type comes from the bytes, never the name or header. Images up to 20 MB, video up to 200 MB, audio up to 30 MB. The same file uploaded twice is stored once. Every asset comes back with a signed URL valid for an hour; that is also how providers fetch it. Uploads are deleted after 60 days, with the outputs.

bash
# upload a frame
curl https://edengen.ai/v1/assets -H "Authorization: Bearer $TAKES_API_KEY"   -H "Content-Type: image/png" -H "X-Filename: frame.png" --data-binary @frame.png

# animate it
curl https://edengen.ai/v1/takes -H "Authorization: Bearer $TAKES_API_KEY" -H "Content-Type: application/json" -H "Prefer: wait=60"   -d '{"model":"kling-3","prompt":"She turns to the window, no music, no cut","resolution":"720p","duration":5,
       "references":[{"asset_id":"as_...","role":"first_frame"}]}'

Roles

RoleMeaning
first_frameThe video starts on this exact image (image-to-video).
last_frameThe video ends on it.
referenceLook, identity or motion to follow, not a fixed frame. Images, videos or audio depending on the model.

What each model accepts

ModelAccepts
Seedance 2.59 reference images, 3 reference videos, 3 audio tracks, a first frame
Seedance 2.09 reference images, 3 reference videos, 3 audio tracks, a first frame, a last frame
Seedance 2.0 Mini4 reference images, a first frame
Kling 3.0a first frame, a last frame
Kling 3.0 Omni7 reference images, 1 reference video, a first frame, a last frame
Veo 3.13 reference images, a first frame, a last frame
Veo 3.1 Fast3 reference images, a first frame, a last frame
MiniMax H34 reference images, 1 reference video, 1 audio track, a first frame
Wan 3.0a first frame
Nano Banana Pro14 reference images
Nano Banana 2.114 reference images
GPT Image 2.510 reference images
Seedream 5.0 Pro14 reference images
Recraft V4.11 reference image
Pika 2.5a first frame
FLUX.1 Schnellno references
FLUX.2 Devno references
Pruna p-imageno references
FastWan QADno references
LTX 2.3 Fasta first frame, a last frame
Pika H3 Flexa first frame, a last frame
Drama 3no references
Eleven Music v2.5no references
Eleven Music v2no references
Seedream 4.0no references
Ideogram Vector 1.0no references
Kling v4no references
HeyGen Video 1no references
FIBO 2no references
FLUX 3 Maxno references
Increase Resolutionno references
Genfill v2no references
Guided Remove Backgroundno references
FIBO Edit 1.5no references
Qwen Image 2.1no references
FLUX 3 Image Kleinno references
FLUX 3 Imageno references
Gemini 3.8 Flash-Lite TTSno references
Gemini 3.8 Flash TTSno references
Gemini 3.5 Transcribeno references
FLUX Video Upscaleno references
Grok Imagine Image 2.0no references
FLUX 3no references
Voice Separationno references
Seed ASR 2.0no references
MAI Voice 2 Flashno references
MAI Voice 2no references
Qwen-Audio-3.0-TTS Plusno references
Qwen-Audio-3.0-TTS Flashno references
Midjourney v8.1no references
Midjourney v7no references
Kling v3 4Kno references
Whisperno references
Reve Imageno references
Seed Audio 1.0no references
Gemini Omni Flashno references
Nano Banana 2 Liteno references
Grok Imagine 1.5no references
Nano Banana 2no references
Grok Imagine Videono references
Wan 2.7 Prono references
Wan 2.7no references
Eleven Dubbingno references
Eleven Voice Changerno references
Eleven Music v1no references
Veo 3.1 Liteno references
Nano Bananano references
Nano Banana 2 (Preview)no references
Nano Banana Pro (Preview)no references
Kling v2.6 Standardno references
Creative Upscaleno references
Creative Upscalerno references
Expandno references
Qwen Image Edit Multiple Anglesno references
Wan 2.6no references
Kling Avatar v2 Standardno references
Kling Avatar v2 Prono references
Kling v2.6 Prono references
Kling O1 Prono references
FLUX 2 Klein 4Bno references
FLUX 2 Prono references
FLUX 2 Flexno references
Hailuo 2.3 Fastno references
Seedance 1.5 Prono references
Seedream 4.5no references
Seedream 5.0 Liteno references
Grok Imagine Imageno references
Qwen Image Maxno references
Qwen Image 2 Prono references
Qwen Image 2no references
Kling v3 Prono references
Kling O1 Standardno references
Wan 2.5no references
ViduQ3 Prono references
ViduQ3 Turbono references
mischief 1no references
Grok Imagine Video 1.5 Liteno references
FLUX Video Editno references
H3 Maxno references
Wan 3.0 Primeno references
Avatar IVno references
FLUX 3 Videono references
Aleph 2.0no references
Gen-4.5no references
Grok Imagine Video 1.5no references
HappyHorse 1.1no references
HappyHorse 1.0no references
Video v3.0 Prono references
Video O1no references
Hailuo 2.3no references
Riverflow V2.6 Prono references
Qwen Image 2.1 Turbono references
Qwen Image 2.1 Prono references
K3 Editno references
Seedream 5.0 Flashno references
Recraft V4.1 Flashno references
Recraft V4 Styles Prono references
Recraft V4 Styles Vectorno references
Recraft V4 Styles Pro Vectorno references
Recraft V4 Stylesno references
Qwen Image 3 Prono references
Qwen Image 3no references
Krea 2 Largeno references
Krea 2 Mediumno references
Krea 2 Medium Turbono references
Riverflow V2.5 Prono references
Riverflow V2.5 Fastno references
Grok Imagine Image Qualityno references
Recraft V4.1 Pro Vectorno references
Recraft V4.1 Vectorno references
Recraft V4.1 Utility Prono references
Recraft V4.1 Utilityno references
Recraft V4.1 Prono references
Recraft V4 Pro Vectorno references
Recraft V4 Vectorno references
Recraft V4 Prono references
Recraft V4no references
Recraft V3no references
Riverflow V2 Prono references
Riverflow V2 Fastno references
FLUX.2 Maxno references
Z-Image Turbono references
Kling 4.0no references
GPT Image 2no references
Eleven Multilingual v2no references
MiniMax Speech 2.8 HDno references
MiniMax Speech 2.8 Turbono references
Sume Remix Motion Transfer8 reference images, 1 reference video
Sume Remix Object Swap8 reference images, 1 reference video
Sume Remix Restyle1 reference video
Sume Remix Lite Motion Transfer8 reference images, 1 reference video
Sume Remix Lite Object Swap8 reference images, 1 reference video
Sume Remove Background1 reference image
Sume Topaz Video - 24 fps1 reference video
Sume Depth Anything V31 reference video
Sume LatentSync - Video Lipsync1 reference video, 1 audio track
Sume Kling Avatar 2 - Talking Image1 reference image, 1 audio track

Ask for something a model cannot take and the request fails before anything is charged, naming what it does accept. Video or audio references route Seedance and MiniMax to their reference-to-video endpoints automatically.

Elements

An element is a named set of up to 9 images: a character, product, location or style. Pass { element_id } and it expands to its images, so the same face or packshot follows you across shots and models. Notes on the element are for you and your agents to reuse in prompts.

python
from takes import Takes
client = Takes(base_url="https://edengen.ai")

faces = [client.assets.upload(p)["id"] for p in ["anna-front.png", "anna-34.png", "anna-full.png"]]
anna = client.elements.create("Anna", faces, kind="character", notes="freckles, green eyes, leather jacket")
take = client.takes.run(model="kling-omni", prompt="Anna walks into the rain, no music",
                        resolution="1080p", duration=5, references=[{"element_id": anna["id"]}])

From agents

MCP tools: takes_create_upload_link (returns a curl command the agent runs locally), takes_import_reference (URL or take), takes_list_references, takes_create_element, and references on takes_create.

References and elements · Docs · Eden Gen