API
Managed visual inference
One API, 110+ production models across video, image, audio, and text. No infrastructure to run.
45M+Images & videos served
110+Models
30Providers
4Modalities
One API. All the top models.
# POST a job
curl -X POST "https://api.hedra.com/v3/models/kling-v3" \
-H "Authorization: Key $HEDRA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"prompt": "Hands resting beside an espresso at a sunlit Roman cafe",
"aspect_ratio": "16:9",
"duration_ms": 5000,
"quality": "pro"
}
}'
submit202
queuesched
runproc
storeasset
observelogs
KLING V3
WAN 3.0
MINIMAX H3
VEO 3.1
SEEDANCE 1.5
LTX-2.3
VIDU Q3
SEEDANCE 2.0
FLUX.3
KLING O3
PIXVERSE V6
HAILUO 2.3Owned, hosted, and routed models behind one integration — one job lifecycle, one bill, routing and scaling handled by the stack that serves our own frontier models.
03Beyond the API
The stack underneath.
The same engine and execution layer we run the API on is becoming available directly. These are conversations today, not self-serve products.
WMR
Our inference engine for diffusion and world models. Runs the API today.
Post-training
Adapt a supported model to your task, then deploy the checkpoint onto the same serving path.
GPU sandboxes
Programmable execution on the sandboxes the API already runs on.
Talk to us about early access
Available through direct engagement. Not yet self-serve.
